The unsustainable exploitation and destruction of forests is a serious environmental concern in the Developing Countries (DCs) of Africa (Bewket, 2003). While rising agricultural yields and rural outmigration have allowed forests to regenerate in some of the richer countries of the temperate world, the poor countries of tropical Africa are fast depleting their forest resources. As estimates indicate, by the early 1990s, the rate of deforestation in this part of the globe was about 29 times the rate of afforestation or/and reforestation (Salih, 1992). If this trend continues, then it would not be too long before forests will be completely removed from the African landscape (Bewket, 2003). One of the main driving forces of the deforestation in these DCs is the pressure from the growing population, which uses forests for fuelwood and domestic energy production purposes (Baohene, 1998). Deforestation has, as is widely recognized, manifold environmental consequences: loss of biodiversity and genetic resources, soil degradation, depletion of water resources, disturbance of microclimates, loss of wildlife resources and impediment to the cycling of carbon to mention a few. Some of these effects are more local while others are global.
Since the 1992 United Nations Conference on Environment and Development in
Rio de Jeneiro, Community-Based Natural Resources Management (CBNRM) has been
increasingly accepted as the privileged solution to the challenges of environmental
conservation in DCs (Virtanen, 2003; Adams and Hulme, 2001; Wily, 1999). The
key message of CBNRM proclaims that if economic development and community participation
are not promoted in conjunction with environmental conservation, then local
populations will have no interest in protecting resources, which will not survive
(Virtanen, 2003). This message has been repeated in scientific publications,
policy documents and popular media until it has turned into a mantra of sustainable
development (Godwin, 2001).
The type of support needed by smallholder farmers varies widely in most of Sub-Saharan Africa. Support may be related to the technology itself, to inputs, production means, infrastructure, marketing and credit, among others. In the case of farmer tree nurseries, specific material inputs required are tree seed and water, inoculum, tools and fencing. These can be classified as hard support. In addition, there is soft support that relates to information, training and backstopping advice. Until recently the provision of agricultural services to smallholder farmers was still largely in the hands of the government sector in East African countries. National agricultural extension systems have largely failed to provide the support needed by smallholder farmers (Oladele et al., 2004). In this regard, the farmer to field staff link has been identified as the weakest one in service delivery. Furthermore, inadequate technologies, high degree of bureaucracy and poor working conditions of field staff are commonly cited as major constraints (Vankatesan and Kampen, 1998). In addition, there is lack of evidence on the contribution of public extension services to the improvement of smallholder farmer livelihoods (Haug, 1999). Overall, current support to smallholder farmers can therefore be considered marginal. As a result, the seed and fertilizer supply sectors as well as the marketing of produce have recently been liberalized as part of economic restructuring in the East African region. These three support components are the most likely targets for privatization because they offer profitable business opportunities, unlike in cases of smallholder farmer training and advisory services, which are likely to remain in the public domain (Vankatesan and Kampen, 1998).
The effectiveness of support systems should generally be measured in terms of what leverage they provide in achieving impact on food security, the creation of wealth and reversing the degradation of the environment. In the context of agroforestry, the extent of tree planting in the agricultural landscape expressed both as area planted to trees as well as numbers of rural households using agroforestry technologies could serve as indicators for assessing impact.
It has been shown that smallholder farmers can play a crucial role in supplying germplasm for development of sustainable land use systems in southern Africa (Boeringer et al., 2003). In order to allow a better assessment of this output, it is important to understand the problems farmers face with respect to nurseries, the support systems that are needed for this achievement and, also the impact of tree planting on farms. Given the declining importance of national extension services in East Africa, it was hypothesized that a number of support agents operating at grass root level together with farmers themselves provide the different support functions needed in farmer nurseries.
The overall objective of this study is to assess the role and functions of farmer nurseries in building natural, human and social capital for the development of sustainable land use systems. A secondary focus is on farmers problems and the role of support systems in facilitating the establishment and operation of farmer nurseries. It further assesses the impact of trees planted from farmer nurseries and discusses policy issues related to germplasm supply for smallholder farmers in the lake Victoria catchment ecosystem.
MATERIALS AND METHODS
Study sites: The study focused on three districts (Kakamega, Busia and Butere-Mumias) in western province of Kenya (Fig. 1). These sites were chosen because they had well established Farmer Field Schools (FFs) that formed part of another project of the Kenya Agricultural Research Institute (KARI). Key land use indicators for the three study areas are given in Table 1. They show marked differences in average population density and hence cropping area available to households.
Survey methods and data analysis: A total of 24 nurseries were sampled
in the three districts, i.e., 6 nurseries from each district between November
2004 and December 2005. During field visits, interviews were conducted with
farmers to collect information on problems encountered and the nature of support
||Selected land use indicators in three study districts of western
|Jaetzold et al. (2007)
|| Study area (Source: Jaetzold et al., 2007)
The survey team asked open-ended questions and gave the respondents
the opportunity to cite up to three major problems associated with their nursery
operations. In the analysis, a Cumulative Weighted Score (CWS) was calculated for each
problem (Ayuk, 1997). Since a problem one farmer ranks as the most important
may be rated as second most important by another farmer or not rated at all,
a simple procedure was developed to determine the relative importance of each
problem. The most important problem mentioned was assigned a value of 5, the
second the value of 3 and the third most important problem the value of 1. The
CWS was then calculated by multiplying the assigned weight by the associated
number of times the problem was mentioned and then summing across the three
most important problems. Thus, for each problem a CWS was calculated and ranked
for all the problems mentioned.
Two types of support systems were assessed. The first one related to material inputs used such as seed, water, inoculum for seedbeds, tools and fencing. The second one, which we call soft support, refers to inputs such as information, training and backstopping advice. For each type of support, providers were recorded. In principle these could be farmers, staff of development organizations, government extension staff or VicRes project staff or any combination of these providers.
A follow-up survey was carried out later in September 2005 to determine the
extent of transplanting of seedlings from nurseries. The number of tree seedlings
transplanted and the field area planted were assessed at the same time. A case
study to gain more insight on the impact of tree planting was also carried out
in the three districts drawing on a total sample of 36 farmers. This sample
was derived from the initial 24 nursery groups that had been sampled. These
nurseries were first stratified according to type, i.e., group or individual
and then according to gender composition, i.e., those being operated by women
or men. Six nurseries were then selected at random from each of the four strata.
One recipient of tree seedlings from each nursery was then randomly selected
from the list of all individuals that had received seedlings, based on farmers
records at the time of distribution. Accessibility of villages during the rains
and availability of individuals resulted finally in interviewing a total of
36 individuals. Data were collected on gender and age of household head, number
of years of household head in formal education, total number of household members
and wealth rank so as to characterize the households. The wealth ranking was
based on a four-tier system recognized by local communities on the basis of
types of asets owned (land, housing, radio, bicycle, tools and livestock). Wealth
ranks with original terminology used in luhya language given in brackets were:
wealthy (mapesa manji), fairly well off (mapesa matititi), poor
(mutakha) and destitute (mutakha kabisa). Project staff used open-ended
questions to address problems experienced and reasons for expanding tree planting.
Regression analysis and independent sample tests (two-tailed t-tests assuming
unequal variances) were carried out to determine the statistical importance
of the postulated relationships.
RESULTS AND DISCUSSION
Nursery problems: The results in Table 2 show that the majority of farmers experienced no problems. It should be noted here that respondents is not perceive seed availability as an option, since VicRes project provided starter germplasm for initiating farmer nurseries. In the absence of such incentives, availability of tree germplasm is often cited as the key constraint to farmer tree planting (Aalbaek, 2001). The most important problems in this study in descending order were pests damaging seedlings, scarcity of water and lack of adequate space for nurseries. All other problems mentioned received CWC of less than 63 (Table 2). It is worthwhile to note that labour ranked consistently low throughout (Table 2). Overall these results on major problems encountered by farmers, show clearly that establishment and management of nurseries was feasible for majority of farming households, even with their limited resources and capacities.
Support systems: Nursery support could be grouped into two categories:
hard inputs such as seeds, water, inoculum, tools and fencing and soft inputs
such as information, training and advice. On hard inputs, farmers themselves
supplied inputs such as water, fencing and tools.
||Cumulative weighted score of three major problems in 24 nurseries
across three study area districts in western Kenya during 2004/2005 growing
This observation demonstrates that providing these inputs to farmers would
be probably counter-productive in making nursery production sustainable. Inoculum
was largely not used in farmers nurseries (80% not using). This low use
of inoculum can be explained by two factors: (1) many farmers still lack knowledge
about use and benefits of inoculating seed beds and (2) relative high costs
for collecting inoculated soil from established stands of trees elsewhere or
non-availability of such sources in some areas. This result demonstrates that
more investments are required in the future in establishing viable grassroot
level seed supply systems in order that VicRes projects role in providing
seed can be phased out as soon as possible, if farmer nurseries are to be made
On soft support, it was noted that support from government extension services
was generally minimal. This is remarkable because this kind of support to smallholder
farmers is considered to be the core business of national extension services,
which should focus on such public good activities that the market place is unlikely
to provide (Van den Ban, 2000). In comparison, support from non-governmental
partners was much higher. Here collaboration between VicRes project and non-governmental
partners also gained importance in disseminating information and providing technical
backstopping and advice. This result, taken in conjunction with VicRes project
was playing a dominant role in nursery training, a trend that is similar to
the one discussed above with respect to support for tree seeds. There is a clear
need for investing more into training of grassroot level trainers in the future
in order to allow international and national research and development organizations
to move into facilitating the wider process of scaling-up of agroforestry instead
of providing large-scale training directly. Overall, the above results on unequal
distribution of support services among providers at the grassroot level are
far from a desired situation where farmers, researchers and extensionists collaborate
equally as social actors within the social practice of agricultural production
(Cornwall et al., 1993 after Haug, 1999).
Regression analysis with nursery seedling output per individual as a dependent variable and the above listed types of support as predictors revealed significant effects in cases of support for seeds (t-value = 0.006) and for advice (t-value < 0.001). Independent sample test showed further that there was a difference (t-value = 0.059) between support provided by VicRes project and that by partners, resulting in an average output per individual of 600 and 220 seedlings, respectively. This may also reflect differences in quality of advice given the variation in levels of human resource capacities among service providers. On a broader scale, this result also points to interdependencies between the building of human and natural capital, the latter being exemplified by nursery output.
Tree transplanting impact: From the total number of nurseries surveyed in the three districts, 400 individuals, 60% being women, received and transplanted a total of 180,000 seedlings. Regression analysis showed that number of seedlings transplanted per individual was explained by nursery type and whether recipients of tree seedlings were group members (adjusted R2 = 0.348). Despite the larger number of trees transplanted from individual nurseries, their overall impact in terms of tree planting at the community and landscape level appeared however small. While nursery location showed no meaningful effect on average number of seedlings transplanted, independent sample tests revealed that home yard nurseries resulted in significantly larger numbers being transplanted (on average 240 seedlings). Since sustainable land use can also be seen as a function of proportion of total area planted to trees, our results clearly demonstrate that counting numbers of individuals transplanting seedlings was not enough in assessing nursery impact but rather that the contribution of different nursery types in transforming land use at a community or watershed level needs also to be taken into account. Furthermore, we also hypothesized that there might be significant interactions between group and individual nurseries in terms of different capital being built, which needs to be assessed in the future. For instance, we observed quite frequently individual nurseries evolving out of group nurseries, the latter providing crucial human start-up capital which later enabled individuals to become more successful in tree transplanting. Such late-benefits from group nurseries need to be evaluated in that respect. This also shows, once again, how the three development functions of building natural resources, human and social capital appear to be closely interrelated with farmer nurseries.
The use of agroforestry on-farm: A total of 36 farmers were interviewed of which, 28 had planted tree seedlings originating from group nurseries and eight from individual nurseries. An interesting aspect here was the fact that 26 individuals out of the 36 sampled had made plans to produce seedlings on their own in the future, while the remainder declared they would continue organizing themselves in groups for raising seedlings. This result underlines the close interrelationships between group and individual nurseries, the former providing a training ground for individuals who operate their own nurseries later. However, there were no significant differences between group and individual nurseries in terms of the size of agroforestry plots being managed eventually by individual farmers.
There were significant relationships between gender and household characteristics
on one hand and nursery impact indicators such as area planted to trees and
number of trees managed by individuals on the other (Table 3
and 4). Multiple regression analysis of agroforestry land
area managed as dependent variable on formal education (t-value = 0.0532) and
nursery location (t-value = 0.010) gave an adjusted R2-value of 0.2101
(F = 0.006). Likewise, number of trees managed could be explained in a meaningful
way (R2 = 0.228, F = 0.017) by age (ns), years in formal education
(t-value = 0.111), gender (t-value = 0.003) and nursery location (t-value =
||Tree planting as affected by household characteristics among
36 smallholder farmers in western Kenya (expressed in P-values from Levens
Test for Equality of variances)
|aNon significant values, all other values significant
||Mean total area planted to trees and number of trees managed
as influenced by selected household (HH) characteristics among 36 smallholder
farmers in western Kenya
The majority of respondents (80.8%) planted trees in mixed systems or
in a combination of mixed and relay cropping systems in line with the main expected
benefit of restoring soil fertility.
Policy issues: A major policy issue emerging from this study is the need for public investment in improving access to water in rural areas. The results of this study show that scarcity of water was the second most important problem facing farmers that wish to establish nurseries. Investments in improving water supply will not only improve human health and increase agricultural production but will also assist in expanding farmer nurseries. Developing and promoting water-harvesting practices can also assist farmers in getting prolonged access to water. Even though this water may not be safe for drinking, it could serve useful functions in other livelihood strategies. The question from an economic standpoint is who should pay for such investments. It can be argued that an increase in expansion of farmer nurseries will translate to increased tree planting that will have direct consequences on the environment. Such benefits extend beyond the farmers fields and thereby justify public investment. The critical role of water in the establishment of farmer nurseries calls for ingenuity in developing appropriate strategies.
Another aspect that needs the attention of policy makers is that of developing a national strategy for germplasm supply. This study, in agreement with the one conducted by Aalbaek (2001), has shown that germplasm supply remains a major constraint for farmer tree planting. The various actors including government extension services, non-governmental organizations, the private sector and international and regional organizations working in this area need to develop national and regional strategies to address germplasm supply issues. Governments should limit their roles to defining an institutional and policy framework that ensures the smooth functioning of farmer oriented germplasms.
Finally, there is a clear need for policies to articulate strategies for providing support in the form of training and advice. Both elements are critical for building human and social capital as the results of this study show. However, a pre-requisite for such policies will be to conduct a cost-benefit analysis of different types of farmer nurseries in the context of smallholder farmer production and livelihood strategies.
Results on support systems for farmer nurseries provided some insights into
current service delivery at the grass root level. Support for material came
largely from single providers with farmers being able to support themselves
for most hard inputs needed. The supply of tree seed was an exemption with the
present VicRes project being the main supplier. This function should however
be seen as a temporary one triggering the scaling-up of agroforestry impact.
There is an urgent need to facilitate the establishment of community-based tree
seed supply and distribution systems involving the private sector and community
based organizations as much as possible in order to make tree planting sustainable
in the region.
We are grateful to the Inter-university Council for East Africa for the financial support to carry out the study reported in this research through funding of the project Transferring Soil Carbon Sequestering (SCS) Best Management Practices (BMP) to Smallholder farmers in Lake Victoria Catchments Ecosystem. We thank our partners and farmers who have been participating in farmer nurseries in one way or another since the 2004/2005 growing season. The three anonymous reviewers are thanked for their comments which have greatly improved the quality of this research. All errors and omissions remain the responsibility of the authors.