3.1 ICRAF's Priority-Setting and Planning in an International Research Context
3.2 Regionalization of ICRAF's Programmes
3.3 Programme Quality - Inputs, Methods and Outputs
This Chapter provides the Panel's assessment of a number of cross-cutting themes related to ICRAF's planning and prioritization activities, their implications for the move towards regionalization and ICRAF's efforts to ensure high quality programmes. These cross cutting issues were identified by the Panel during the first phase of the EPMR, the field visits and the early days of the main phase. The themes address issues and questions raised in Chapter 2. Specifically, they are as follows:
· How does ICRAF set priorities and plan for its research and development agenda? Is the process adequate in terms of linking goals to actions? What are some of the ways in which the process might be improved? (see Section 3.1)· ICRAF has moved recently to a regional model where much of the activity is driven by regional needs and opportunities, but within the overall context of the ICRAF strategy, goals, and objectives; how does this regional model work? Is it an appropriate way for ICRAF to organize its activities? Does this approach provide for adequate linkage among regions and between regions and the global programmes at headquarters? What improvements can be suggested? (see Section 3.2)
· A key question to be addressed in any review of CGIAR Centres is the quality of their programmes - which the Panel interprets broadly to include quality of inputs, problem formulation, methods and implementation, and outputs. Quality is a key element to consider in planning. (see Section 3.3).
The 1993 EPMR recommended that ICRAF "develop a more transparent and systematic research programme planning process, including priority setting, monitoring and evaluation..." ICRAF responded by initiating an iterative and participatory process in January 1996, so as to streamline the preparations for the next (now current) Medium-Term Plan accordingly.
As stated in this MTP for the period 1998-2000, ICRAF attempts to set its research and development agenda in response to the global challenges of the next 25 years, and in accordance with the overall CGIAR research agenda. Thus ICRAF's activities are focussed on contributing to improvements in human welfare and environmental resilience through research and development in agroforestry land-use practices (mission statement). Implementation of the planned activities involves an integrated natural resources management approach applied through regional programmes, with backstopping from the global programmes at headquarters, as outlined in Section 2.5.
The annual planning process is designed to start in the regions with meetings attended by numerous partners and other stakeholders who bring forward their local priorities and propose implementation responsibilities. Programme Leaders from headquarters who are overseeing or proposing research in the region also attend, and their presence and typically strong negotiating stance are thought to contravene a focus on overly localized research. Programme staff at headquarters hold their own planning meetings, but plans are not finalized until both regional and programme staff meet for the Annual Programme Review (APR), a two-week review and planning marathon held each September. Results of the APR may necessitate adjustments to both regional and programme plans made earlier. The outcome of this iterative planning process is incorporated into ICRAF's Programme of Work and Budget (PWB) and presented for approval of the Board of Trustees at their December meeting. Any needed final adjustments are made at a Management Retreat in January, attended by all senior and middle managers. The Panel commends the design of this process but has observed that funding shortages in the Regional Offices sometimes precludes the scheduling of formal annual meetings with partners.
According to ICRAF's Strategic Plan, the Centre is expected to continue its activities on the humid, sub-humid and semi-arid tropics, "areas where essential increases in food and fibre production must be obtained in ways that mitigate the major ecological threats of deforestation and massive land depletion." The Plan included criteria for the selection of regional sites, among them: a) the potential contribution that agroforestry technologies could make to specific ecoregions and to priority land-use systems; b) the constraints and potentials facing rural households; c) the interest demonstrated by national institutions and donors; and d) the comparative advantage of ICRAF to undertake specific research and dissemination activities in the region. The ASB process for the selection of programme benchmark sites supplemented these criteria. The Strategic Plan provided a starting point for more detailed planning that culminated in subsequent Medium-Term Plans.
In the Panel's view, the current MTP for the period 1998-2000 is a useful document for relatively short-term planning-but primarily at the programme level. The Medium-Term Plan does not substitute for a Centre-wide research plan (except as an aggregation of discrete projects), and it does not contain explicit research plans for individual regions, although it lists the general objectives of the research for each. As a result, Regional Offices work primarily from an annual plan presented in the Programme of Work and Budget. They do not have strategic plans within the Centre-wide document, nor has a Centre-wide strategic document been developed subsequent to 1993.
Some staff have observed to the Panel that ICRAF is more proficient at short-term, annual planning than at planning over a longer term. Ideally, of course, an annual plan should derive from a research plan of three to five years duration that, in turn, should be based on a Strategic Plan covering a full decade or more. To be sure, even the Strategic Plan should be revisited from time to time as conditions change, opportunities arise and knowledge of needs and solutions increases. The Panel is not certain that ICRAF's planning processes adequately comprise consideration of these several time frames nor that they adequately provide for iteratively considering global issues in the light of regional requirements.
The Panel was pleased to note that the Director of Research has formed a small team to design a priority setting process that will employ sophisticated systems methodologies to be able to justify what work is done where on the basis of an explicit and quantitative assessment. The methodology is designed to yield priorities, objectives, consensus and understanding, new opportunities and a cohesive framework for planning, impact projection and impact analysis. It will be interesting to learn what the group comes up with and whether the results suggest a change in direction at any level, regional or global. Figure 3.1 shows a useful event chronology for NRM research being further developed by the team.
Fig. 3.1: An event chronology for NRM research
The development of this process and its application will, at best, take some time. The Panel considers it important, however, that ICRAF review its annual planning process and its relationship to longer-term plans. A schematic strategic planning framework for guiding this process at the regional level is given below in Figure 3.2.
Fig. 3.2: Strategic Planning Framework for Regional Activity
While all regions will be different in terms of how they provide an appropriate mix of support to regional partners, the three basic elements are needed - work on policy context (primarily Programme 1 input), direct agroforestry research and development (all Programmes) and training and capacity strengthening (all Programmes, with a focus on Programme 5). This planning framework fully complements ICRAF's five research pillars and highlights the strategic dimensions of the Centre's work.
To help operationalize this framework, the Panel recommends that:
ICRAF review and update its 1993 Strategic Plan, develop both strategic and medium-term plans for each of the Programmes and Regional Offices and use these as the basis for annual planning.
3.2.1 ICRAF Strategy for the Regions
3.2.2 Region and Headquarters Interactions and Coordination
3.2.3 ICRAF's Regional Role in the CGIAR Systemwide Programmes
The six regions of ICRAF's organisational matrix are geographically diverse, with four in Africa, one in Latin America and one in Southeast Asia. The level of activity, funding basis and mix of programme activity vary considerably between regions. In fact, as mentioned earlier, the ICRAF HULWA region will close down due to lack of funding support and the recognition that ICRAF's comparative advantage in HULWA was in the area of tree domestication and how to integrate trees into the farming system. IITA as the regional coordinator for ASB and the eco-regional programme, EPHTA, is responsible for other aspects such as characterization, farming systems, crop improvement, policy, etc. The matrix management system is still evolving with the recent transfer of considerable decision making authority and fund raising, financial management and reporting/evaluation responsibilities to the regional coordinators.
The Panel looked at several questions related to the rapidly evolving regionalization of ICRAF's programmes. One relates to the overall strategy that ICRAF management has adopted for the regional programmes. A second question relates to the relations between headquarters and the regional teams - both actual and those intended for the future when the system is fully developed. Third, there is the question of whether or not the regional approach is advantageous for ICRAF in its activities in the Systemwide programmes established under the CGIAR umbrella.
ICRAF's current strategy incorporates the idea that each region should have some activity from each global Programme. While this might seem desirable in order to fulfil ICRAF's mission, the Panel believes that funding limitations, as well as the great differences among regions, including different priorities, will make this difficult to achieve, and perhaps not desirable. The differences among the regions are such that one or more programme elements may not have top priority. Instead heavier concentration in a few programmes may be the most appropriate approach for a given region, as is currently the case.
ICRAF has developed several approaches to the challenge of staffing the regions appropriately. For example, in the SEA regional model, three regional IRS staff located in one sub-centre relate to three different global Programmes; and the activities undertaken in the region relate mainly to those three. The IRS staff provide a critical mass, a defined problem and a methodology for developing an approach to solving it. Along with functioning logistical support, this has attracted scientists from ARIs and from NARS to work with the ICRAF staff, operating from their offices on the CIFOR campus. This leveraging of ICRAF's investment enables establishment of viable problem-oriented teams, often with an expanded discipline base paid for by others.
The SEA region provides a good model for the others, although the Panel notes that most of ICRAF's regions do not yet have the disciplinary mix to operate in the same manner. However, they can aim towards the same general type of synergism and collaborative arrangements demonstrated by the SEA programme.
Appropriate location of the regional programmes is important. Where the regional headquarters is located in one of the NARS research stations, this provides ICRAF with the opportunity to support the development of that station for the NARS, through support in upgrading both building and field facilities. There is a tendency in such circumstances to then regard the building (and the station in some cases) as ICRAF's, and this is unfortunate. It is at this level that the most realistic model for cooperation (and capacity strengthening) with the NARS in general should be set. The culture of interaction with colleagues who are not ICRAF staff is vital to the success of ICRAF's mission. This issue is also dealt with in Chapter 4.
Another pattern of operation in the regions is the posting of individual IRS in-country, sometimes in isolation. This necessitates a different modis operandi for ICRAF. In such circumstances, the programme allegiance of the staff member becomes less important than the allegiance to ICRAF, because she or he is in a sense representing ICRAF as a whole. The position requires someone who is an integrator and a facilitator, since what they can achieve in the way of research on their own often is limited. Nethertheless, ICRAF has shown how well this can operate, e.g., in Thailand where the ICRAF's senior scientist (a policy analyst) has developed a network of collaborators covering both socio-economic and biophysical aspects of the problem of managing a mountainous forest margin and extensive associated watershed. It is also in this situation that the regional staff person requires good support from headquarters Programmes in the technical aspects of the work, so that scientific standards are maintained. As indicated elsewhere, the Panel cautions ICRAF about outposting scientists alone in a country or region.
A further extension/development dimension has been added in all regions with ICRAF's move towards farmer participatory research as the pathway to test the robustness of the research findings and to facilitate dissemination. Regional programmes have different levels of co-operation with the NARS and NGOs in this process; but it is at this level that the dissemination and rural development starts to take a sustainable, concrete dimension. ICRAF regions need to keep uppermost in the scheduling of their activities the recognition that, at the development end of the spectrum, ICRAF's main role is to test the effectiveness of different methods of extension of NRM research rather than to be the main extension agent per se. Having once developed techniques that are effective, then ICRAF's role reverts to one of training the trainers in those techniques, as part of an agreed strategy with their NARS and NGO partners.
Maintaining the balance between these research and development components at the regional and headquarters levels will be a dynamic process. The effectiveness of the regional programmes will thus depend a great deal on the capacity and management skills of the Regional Coordinators and the way global Programmes make their input to regional activity, both in terms of collaboration in planning research and in providing technical advice and analytical support from the headquarters laboratories. These interactions are very much in an evolutionary phase at ICRAF. The strategy and research plan, combined with funding, have an overarching role in determining the emphasis that regional programmes place have on research, institutional support and extension to farmers. The present funding emphasis is towards development activity, so that regional programmes are generally operating at the applied end of the research spectrum with an increasing emphasis on farmer participatory research and exploration of ways to extend new technologies to farmers.
ICRAF regional programmes already have a wealth of experience with different aspects of ICRAF's overall programme, including both the research side and the development side. In the near future, that experience needs to be drawn upon in a systematic way to ensure that the process of regionalization is cost-effectively implemented.
Thus, the Panel recommends that:
ICRAF do a systematic analysis within the year of the factors associated with successful and unsuccessful operation in the regions that draws out the lessons for an updated set of guidelines for management of the regional programmes.
The Panel believes that such an exercise, which should include some outside expertise to provide an objective perspective on the headquarters/regional linkages, would be of great benefit to ICRAF. While it might be argued that it is too early for such an assessment, the counter argument is that an early one provides feedback while the regional programmes still are adjusting to the new authority and responsibilities given them, and thus are less set in their ways.
The ways in which the Research and Development Divisions at headquarters support the regional activity will be important in ensuring that ICRAF maintains a cohesive vision and "programme of action" rather than operating through a series of autonomous, regional mini-ICRAFs. This is in no way a criticism of the principle of decentralization to the regions of management functions. Rather, it points to a challenge to ICRAF's various levels of management and operations to improve their functioning in the new structure.
The fund raising process, which has now been moved out much more to the regions, requires good communication with ICRAF headquarters to ensure coordination of the project genesis/development cycles so that regions are not competing for donor funds in a manner that distorts ICRAF's overall agenda. ICRAF has moved the Projects Office from the DG's office to the FINAD Division. It assists in processing project proposals, but mechanisms will be needed to efficiently keep management informed of the status of proposals so that they can coordinate their representations to donors about funding and reduce any tendency to develop major imbalances in funding between regions. Obviously, some donors will prefer to work with particular regions and have their own priorities on the type of project they wish to fund. However, the more they feel part of the ICRAF planning process and use the proposed research plan as a guiding document to help decide what is included or excluded, the more likely a mutually acceptable outcome will emerge. The "Agroforestry for sustainable rural development in the Zambesi basin, project management plan," provides a good model of the way this donor-ICRAF Regional Coordinator-senior management interaction may take place.
The evolution of the regional programme model has been rapid and it is much to ICRAF's credit that there have been relatively few areas for concern thus far. What has been learnt so far is that it is necessary to have a critical mass of staff and support funding if the region is to function as a true (research and development) programme. The lesson from the headquarters side is that the time commitment for headquarters based staff is considerable and there is a challenge in servicing such widespread locations and coping with the travel involved.
Another aspect of ICRAF's regional activity is its role in the CGIAR Systemwide ASB and AHI programmes. The ASB consortium (AARD-Indonesia, EMBRAPPA - Brazil, IRAD - Cameroon, CIAT, CIFOR, IFPRI, IITA, IFDC, TSBF, ICRAF, ten advanced research institutes and five other major national programme partners as well as 29 other national research institutes, 18 universities in collaborating countries, 43 NGO's) developed as a multi-institutional multi-national project funded by the Global Environment Facility of the UN and UNDP from 1994 with a one-year cycle and a mandate "to work towards mitigating tropical deforestation, land depletion and rural poverty through improved agroforestry systems" through research on both technical interventions and policy options. ICRAF coordinates this global activity through a half-time senior staff member who also works with the Cornell International Institute for Food, Agriculture and Development on related activity, and a full-time assistant coordinator. ICRAF's regional work in Latin America, Southeast Asia and Humid West Africa, as well as global research at headquarters, represent ICRAF's direct contribution to the ASB agenda.
Part of ICRAF's work in the Eastern and Central Africa (ECA) region is a direct contribution to the ecoregional AHI programme, where the main concern is "improving Natural Resource Management within the context of improving agricultural productivity in the highlands." The AHI seems poised to make a substantial and beneficial impact on the way research is conducted in the region. The synergies accruing from the collaborations with the NARS and ICRAF and other international Centres, and from concentration of efforts on well chosen representative sites, will do much to improve the capacity of the NARS to deliver outcomes that will have impact at the farmer level. Although much of phase 1 was spent in planning, this was necessary for the sustainability of the AHI itself. The AHI programme was comprehensively and favourably reviewed for ASARECA and the donors supporting AHI in 1996.
In sum, both AHI and ASB provide good examples of the facilitating role ICRAF can play in institutional strengthening of NARS, while at the same time conducting research producing IPGs. The gradual transfer of responsibility for management of the AHI and ASB to the NARS is occurring in a carefully planned manner that says much for the skills of the coordinators. It is to be hoped that donors can see that the stage is now set for very productive outcomes from the NARS, supported by the technology and know-how available from the IARCs involved.
ICRAF has transformed itself in the last five years, establishing its mark as a scientific research institution able to play a significant global role. A key feature of ICRAF is that it deals with the issues surrounding agroforestry in a holistic manner. The recognition of the pivotal role to be played by the social sciences in setting priorities, in developing strategies and in the dissemination process has been central to this. At the same time ICRAF has developed capacity in understanding processes at both the biophysical and socio-economic level and has embarked on placing the knowledge gained in a systems context and taking the technologies developed as soon as possible to on-farm tests. ICRAF is to be applauded for putting much effort into the testing of the farmer participatory research model as a method to combine research and extension. Much energy, commitment and goodwill on the part of the NARS collaborators have gone into this effort. With such a rapid development there are bound to be areas where some change in approach or style of interaction are needed, but we believe ICRAF has the capacity and the will to undertake these.
Time Pressures and Quality of Output: One of the casualties of this rapid development has been the time available to develop genuinely collaborative research projects within the Centre. Not surprisingly this has been less of an issue for some of the regions where the small number of IRS has promoted team building and a problem-oriented approach. The annual review and planning meetings, while charged with energy and good intentions are no substitute for the need to reflect and discuss together at a neutral forum in a less pressured atmosphere, to define the problem and the approach to be used in its solution. This process should be undertaken in the context of the overall strategy for ICRAF to discharge its mandate as discussed in Chapter 3.1. ICRAF needs to recognize that this reflective and interactive part of the planning process is not time wasted but is a means to focus the considerable intellectual capital available in the Centre on a collective outcome.
The recent CCER for Programme 3 commented on the separateness of the ICRAF Programmes in general. This is at an intellectual level and is a separate issue to the atmosphere of good will and camaraderie that emanates from the Centre. The domestication of trees for example will need a great deal of input from Programme 1 in assessing the routes to market development, Programme 3 for inputs in ways to establish the plants and maximize their growth, Programme 4 for putting the use of these trees in a farming systems context and evaluating farmers abilities to cope with these new inputs, and Programme 5 for helping to disseminate the new knowledge.
One of the main competing activities for this shared time in problem solving is the need to be continually seeking new restricted funds. ICRAF must re-examine its approach for managing this process so that those with acumen in this area are assigned the task rather than it being everyone's task. The individual scientist needs to be involved in the discussions in setting up the issue, making some initial contacts and in helping prepare the project document but then needs to be gradually relieved of the main responsibility for taking the project proposal further. Part of this process of reordering time allocation should involve a much more rational and ordered approach than at the present for deciding on what collaborations to pursue in projects and this is discussed in Sections 3.1 and 4.2.2.
Researchers and Facilities - Quality Issues: The staff at ICRAF are both committed to their duties and generally of high calibre. The skills required and the type of personality that best carries out certain activities vary, but ICRAF has in general got this mix right. All staff need to be cognizant of ICRAF's role in capacity strengthening and the ways that this can be carried out. There is a continuing need to be aware of the benefits of genuine collaboration, particularly with the NARS.
The Panel recommends that:
all IRS and senior nationally-recruited staff take part in regular training courses on creating working partnerships, on working in teams and on recognizing the influence of personality and of cultural differences in relationships.
The new facilities at ICRAF headquarters are very good by any standard and in part lead to the congenial atmosphere for research. However the very excellence of this input to the programmes needs to be used carefully in building up collaborations with the NARS. The atmosphere in these relationships needs to be one of genuine sharing in the good fortune to be able to use such good facilities on an equal basis. For this reason ICRAF needs to carefully consider its policy for charging for use of the facilities at headquarters by NARS as some consideration should be given to the issue of the volume of throughput achieved and achievable, and the value put on goodwill developed by sharing the facilities at an affordable cost. There is also a good opportunity for ICRAF to use its very good soil chemistry research capability in support of NARS through the TSBF network.
One of the major inputs to programme quality is the funding available for capital equipment and operating costs. When much of this input as well as staff costs need to be met from unrestricted funds, then the competition for these funds may mean that the Programme may not obtain enough resources to carry out its Centre-agreed research agenda. The Centre needs to find ways of obtaining restricted funds to support Programme 3, which is the most affected by the lack of such funds at present. One way to do this is to draw funds from the restricted projects of other Programmes that it supports with necessary knowledge, often gained from Programme 3 research. They should also seek funds from projects with a development emphasis where some supporting research is still needed. Any development project dealing with nutrient replenishment or with watershed management would obviously fall in this category. This process knowledge is essential if ICRAF is to use research to guide its activities rather than to rely on empirical testing of ideas.
Quality of the Methods Used: The methodologies that ICRAF are using in its research have generally been at the cutting edge. The research on plant/water relations and the interaction of organic/inorganic fertilizer interaction, on systems evaluation of the effects of agroforestry interventions, for example, has laid the groundwork for a scientifically rigorous analysis of the way to develop agroforestry systems for different regional environments. There is a need to see that the level of quality control exercised at headquarters flows through to the sometimes less experienced staff in the regions to ensure that the research is necessary, is using appropriate methodologies, is analyzed appropriately and that the implications of the results are understood and are then transmitted to the rest of the ICRAF community. The Panel found a very few instances where more rigour may have been warranted. ICRAF has early on in this five-year period undertaken careful reviews of some of its earlier research in the regions and has shown a commendable ruthlessness in stopping experiments that may have been poorly designed.
ICRAF has shown a capacity to move into new research areas once a line of enquiry has achieved the required results. For example, in the area of plant/water relations between trees and crops, ICRAF developed techniques for examining water use by trees and from this was able to show where, and how, competition for water reduces crop yields and where this overrides any benefits from mixtures of trees and crops that otherwise may have accrued. In the process, inexpensive and robust instruments were developed for measuring water flow through roots.
Following this line of research on the use of water at a field level, the issue of water use at the watershed level was then addressed, and ICRAF scientists from Programmes 1, 3 and 4 have developed techniques for addressing the effects of variation in the landuse at the landscape level. ICRAF is thus well placed to take the lead role in the coordination of research to be undertaken under the Systemwide SWIM initiative project on Improved Water Utilization in a Watershed Perspective.
ICRAF has developed a systems approach to the replenishment of soil fertility. It incorporates research results on changes in soil fertility through various means, including improved fallows, biomass additions and inorganic P fertiliser additions; erosion control measures; planting regimes; and on-farm experimentation as a vehicle for testing methodologies and disseminating ideas. Socioeconomic factors in farmer uptake of the practices are being analysed at several locations in Kenya, Malawi and Zambia in collaboration with NARS and NGOs. ICRAF's catalytic role in this rapidly expanding endeavour has been very commendable and indicates the role ICRAF could play in support of larger national development projects on soil nutrient replenishment or "capitalisation".
An area where ICRAF's methodological approach can be improved is in the development of animal fodder for small scale dairies. There is a need to integrate this research and dissemination activity much more closely with other aspects of animal management such as feed nutritional quality and the role of animals in the production system. Closer collaboration with ILRI in designing and operating such projects would help to do this. It is not much use providing fodder for dairy cows if the feed is of poor nutritive value, if the animals are in poor health because vaccines are not available, if they cannot be gotten in calf, etc.
ICRAF has developed methodologies for the collection of germplasm of trees that have a potential role in agroforestry. Farmers' evaluation of the "value" of trees that they use has helped set priorities for the collections undertaken. ICRAF's multinational Advisory Committee on Genetic Resource Activities plays an important role in reviewing and helping plan the activities of Programme 2 in this area. There is need for Programme 2 to reconsider its logical framework to encompass the broader issues of marketing in its concept of domestication. ICRAF has close links with IPGRI and ILRI in this genetic resources area and with ILRI has undertaken joint collections of Sesbania spp. that can now be housed at ICRAF. They have also undertaken collaborative research on molecular genetic analysis of fodder legumes and training in germplasm collection techniques.
There is an increasing world-wide interest in carbon sequestration as a major IPG issue, and ICRAF's development of methods to model the effects of land use and agroforestry interventions in the humid tropics on the store of carbon in the soil has made a major contribution to the way in which soil as a sink for carbon can be factored into global models of carbon pool sizes. Recent ASB consortium findings demonstrate that the carbon sequestration potential of the most promising alternatives to slash and burn agriculture in the humid tropics are those involving agroforestry trees; most of the carbon sequestered is in the above-ground biomass. By comparison, in cropping systems at the forest margins, most of the carbon sequestered is in the soil. ICRAF can capitalize on its ability to use the methods developed under ASB to make a contribution to IPG activities but needs to carefully assess where it has a comparative advantage and the resources to follow through such lines of research.
The policy research undertaken by ICRAF in Indonesia on the management of the forest margins has been ground-breaking in that it showed how analysis incorporating biophysical and socio-economic information could lead to fundamental changes in the ways governments view forest lands and the role of local people in preserving biodiversity and existing agroforests.
Quality Assurance: ICRAF has developed a procedure for ensuring the quality of its outputs. The global Programme Leaders provide technical backstopping and advice for the regional programmes. Reviews at the APR, by the Director of Research, by Programme Leaders and peer reviews for publications are all part of the process. Furthermore staff assessments also deal with issues of the quality of the scientific output.
In addition, ICRAF has commissioned ten Centre-Commissioned External Reviews (CCERs) in the last five years. These were targeted at a range of issues, and only five deal with overall Programme, Division or region performance. The reviews are of variable quality and hence usefulness. Special attention should be paid to the TORs for the CCERs. It would also be helpful to use reviewers that have some familiarity with the CGIAR system and ICRAF's role in it. It is not helpful when CCER's go beyond their TORs because this reduces the time spent on programme issues per se and hence the quality of the output. The recent CCER for Programme 3 is a case in point.
Disciplinary Coverage: There are still some gaps in the Programmes in terms of their discipline coverage. For example in Programme 2 there is a need for more input into nursery propagation and establishment techniques for the trees under domestication. In Programme 3, there is a need to measure the amounts of nitrogen being fixed by the legumes used in the agroforestry systems ICRAF is developing and to undertake research on ways to maximize this, particularly if the germplasm has been transferred from another location. This could be undertaken in-house and/or through projects developed in collaboration with NARS (such as KEFRI), universities and ARIs with capability in this area. ICRAF has no in-house expertise in pest management, and Mesoplatys beetle and nematode attacks on Sesbania used in improved fallows have been identified as a problem already. With expanded areas of plantings of agroforest trees/shrubs, pest problems are likely to appear. Organizations such as ICIPE and CABI based in Nairobi have the expertise required to work with ICRAF on these issues. In the Development Division there is a need for additional expertise in the area of extension methodology to be able to set up different approaches and test their effectiveness (see also Chapter 4). ICRAF, with its laudable commitment to taking the process of research into the development end of the RtoD continuum, is ideally placed to undertake some groundbreaking and much needed research in this area.
The Panel recommends that ICRAF further develop the research and training needed in the area of vegetative propagation, nursery management and establishment.
The Panel suggests that ICRAF develop research capability in the measurement and enhancement of nitrogen fixation of trees in agroforestry systems.
The Panel suggests that ICRAF work with ICIPE and CABI in the area of pest control in agroforestry systems rather than develop in-house capability.
ICRAF's programme on domesticating trees that produce high value products needs to be supported by careful analysis of the marketing aspects of such products in order to define the ways in which poor farmers will benefit now and continue to do so. Expertise to undertake such appraisals is not presently available in ICRAF, and the strategic planning process in this area is thus incomplete.
The post-harvest aspects of the products from domesticated trees is another area where disciplinary strength is lacking. Since these are usually product specific, such expertise should probably be obtained through consultancies or from NARS who may be able to address this issue.
In Sections 3.1 and 3.2, the need for policy research as an important part of the strategic planning process has been addressed. This disciplinary support is not currently available in all regions. In the Latin America Regional Programme (LARP) a closer link with the CIAT staff member posted under the ASB programme at Pucallpa, who has expertise in policy research, may be a way to provide the necessary support.
ICRAF otherwise has a good mix of disciplines and skills amongst its IRS and should explore ways of complementing its needs by closer collaboration with the NARS or by secondments from ARIs. It should be possible to increase the level of short-term secondments from the NARS with salaries paid for by the NARS and/or to expand the visiting scientist or fellowship programme. The experience with this under AHI indicates the potential. The other way to foster genuine collaboration with the NARS is to commission some work using specific expertise and facilities that the NARS have as part of a collaborative programme as in the model developed by ASB and AHI. The AHI programme, for instance has set up the culture for genuine cooperation between institutes. ICRAF with ASARECA needs to vigorously pursue funding for AHI to bring this up to the rather modest budgeted level.
A significant gap in the expertise base is in the area of extension methodology. ICRAF needs to undertake research on the ways to disseminate its technologies to farmers and to evaluate those being used, particularly those in the pilot projects. A recommendation on this is made in Section 4.1.2. This process will help ICRAF to determine how best to undertake its development activity. For example, in the SARP, should more emphasis be placed on the train-the-trainer programme (in order to build up the capacity of the Malawian extension services and NGO's) to undertake on-farm dissemination of improved fallow technologies or further develop the process of on-farm experimentation as the entré to farmers.
Potential for External Interactions to Further Quality: There are four international institutes (ICRAF, ILRI, TSBF and ICIPE) as well as regional projects such as RELMA and the collaborative GTZ-ITFS Project with headquarters located in Nairobi, all with an interest in socioeconomic and spatial data bases. ICRAF is in a good position to take the lead on making whatever information is available collectively from surveys etc. and to develop a network that keeps in touch about further raw data collection and GIS outputs that have used these, so that duplication is minimized. One of the quality outputs from having such a large scientific community in Nairobi, in the NARS, at the universities and in the international organizations is the opportunity to undertake a dialogue through seminars and workshops on cross-cutting issues. This is a challenge for all the organizations to better use this intellectual capital.
With the increasing interest in such major public goods activities as carbon sequestration and watershed management, ICRAF needs to very carefully assess whether it has a comparative advantage and the resources to follow through such lines of research. ICRAF could play a coordinating role in such endeavours if it is well placed from its physical location to do so.
One of the aspects of output that needs further consideration in ICRAF is the sharing of responsibility for further development of maturing technologies by the NARS. The pilot project at EMBU is a particular case to be re-examined in this regard, and the relationship with KEFRI at Maseno in terms of station management on a day to day basis also needs attention so that the allocation of shared responsibility is consistent with ICRAF's comparative advantage. In Southern Africa Regional Programme (SARP) in Malawi and Zambia it needs to be recognized that the project is operating on a government station and that it is not just an ICRAF project as the signs would tell us (and the attitude of some of the staff) but rather a collaboration with often several ministries. ICRAF also needs to ensure that its seconded staff and NARS colleagues are not differentially affected in situations where operating funds are limited.
Quality of Publications: ICRAF has made a commendable effort to publish its results in refereed scientific journals, conference proceedings, book chapters, reports and more popular press articles. There has been a dramatic increase in peer reviewed articles from 22 in 1993 to 109 in 1997 and a change in articles per senior scientist from 0.2 to 1.1 per year. As detailed in Chapter 2, ICRAF's output is substantial. There has also been a rapid increase since 1994 in the number of ICRAF's papers cited, based on the Science Citation Index. It increased from five in 1994 to over a hundred in 1997. ICRAF has published 251 papers in refereed journals from 1993 to 1997, with 33 percent of these co-authored with NARS partners and 18 percent with partners in the North. In 45 percent of the joint papers, a NARS collaborator was first author, whereas in the prior five-year period the figure was only 4 percent of the 11 percent of co-authored papers. This is a dramatic indication of ICRAF's commitment to undertaking research with NARS scientists. ICRAF's articles have been published in 50 scientific journals, and this has helped them reach a wide audience.
During the five-year period to 1997, ICRAF published 695 articles and reports including 84 book chapters. Again, authors from the NARS contributed to 26 percent of all the articles published by ICRAF, while authors from the north contributed to 15 percent. If one uses the index of the ratio of total publications to refereed journal articles, ICRAF is as low as 2.8. These publications have made a very significant contribution to the science and practice of agroforestry, laying the theoretical groundwork for the benefits or otherwise of the use of trees on farms. The recent work on systems modeling of on-farm effects of agroforestry that goes across programmes is a significant advance in the use of this knowledge to predict the role of agroforestry across environments and farmer practice.
Programmes 3 and 4, with more staff in place over the period, have each published more than 70 papers in refereed journals during this period. Programmes 1, 2 and 5 have published 45, 53 and 7 respectively. Articles published in conference proceedings range from 50 in Programme 4 to 32 in Programme 3. Programme 5 has produced a series of training modules being used in 60 universities in Africa, as well as a four-module course on Agroforestry Research for Development/Integrated Land Use and another on Experimental Design. A CD Rom data-base on agroforestry has been produced, and wherever possible AFRENA NARS have been set up with E-mail connections. Sixteen slide modules, a video and other promotional material have also been produced. These are high quality outputs.
In conclusion, the move into examining the effects of spatial diversity as well as structural diversity in ecosystems and landscapes, coupled with satellite imaging to map and monitor changes in the landscape, will bring together Programmes 1, 3 and 4 into developing decision support systems and integrated national resource management and keep ICRAF at the cutting edge of this rapidly developing integrative scientific endeavour. The on-farm trials are testing a new approach to undertaking adaptive research and at the same time are assessing dissemination methodology.
ICRAF has the challenge to ensure that its excellent strategic research programme develops vigorously and keeps in step with the more applied on-farm research. If ICRAF is to remain viable as a leading research institution, it needs both.
