Forest genetic resources and farmers’ tree domestication Presented at regional workshop on Learning agrobiodiodiversity: options for universities in Sub-Saharan Africa - Nairobi January 2009 Ramni Jamnadass, Ian Dawson, Roger Leakey, Roeland Kindt Jonathan Muriuki, Jan Beniest Tony Simons

Background issues Domestication defined Tree genetic diversity ICRAF Resources

What are the big social, economic and environmental issues in the world? POVERTY DeforestationOvergrazing ECOSYSTEM DEGRADATION AND SOIL EROSION Unsustainable Breakdown of nutrient cycling Cropping and loss of soil fertility/structure LOSS OF BIODIVERSITY Loss of income from wildlife and plants Breakdown of ecosystem function Increased transport Loss of income from crops of nitrates to water table Loss of crop yields MalnutritionIncreased health risks DECLINING LIVELIHOODS AIDS External issues

How can we address these issues? Need to simultaneously restore: biological resources and natural capital (soil fertility, water, forests, etc), livelihoods (nutrition, health, culture, equity, income), agroecological processes (nutrient and water cycles, pest and disease control, etc.). Agroforestry can contribute to these objectives. There is not a simple answer.

Agroforestry promotes agroecological succession Natural ecosystems progress from a ‘pioneer’ stage to ecological maturity. Likewise each phase of an agroecological succession will be more biodiverse, as the planted trees, crops (the planned biodiversity) and introduction of livestock/poultry/fish etc are enriched by the unplanned biodiversity, made up of all those organisms, above and below ground, that find niches to fill among the plants and animals.

Agroforestry contributes towards diversification to create mature or ‘climax’ agroecosystems

Agroforestry promotes Multifunctional agriculture 1. Agroforestry is the integration of trees into the farming system that provide a wide range of products and environmental services: to diversify the farm, to restore ecological services and environmental resilience. 2. It is developing this mixed farming system so that it becomes more productive and generates income and employment opportunities, so that the household livelihoods are restored. This can be achieved by domesticating trees for agroforestry.

Promoting agroecological function Improved soil structure and organic matter management Enhanced nutrient cycling – soil invertebrates, saprophytic and symbiotic fungal and bacterial associations (BNF) Improved water use efficiency More effective pollination Enhanced food chains / life cycles - reduced pest, disease and weed outbreaks (scale dependent) Carbon / trace gas sequestration

Agroforestry… …is uniquely suited to address the requirement for increased food security and biomass resources, and the need to sustainably manage agricultural landscapes for the critical ecosystem services they provide.

Trees are found in three places Natural Forest Plantations On farms

Where are the forests going?

Future of Trees is on Farms Need for domestication? evolution has created 60,000 tree species for thousands of years human extracted what they needed from the forest now population exceeds extractive capacity popn was 1 billion, today 6 billion - original global forest cover 70%, now 26% most tree species are wild Need for domesticationNeed for domestication

Tree breeding practices largely based on recurrent selection for additively inherited traits (P= G+E+GE) heritability (h 2 ) = G A /P (0.1 low h 2, 0.4 high h 2 ) high selection intensities (1 in 100,000 trees) genetic gain of 10-25% for tree volume more recently breeding for hybrid vigour (tropical pines) also recently clonal forestry, esp. rooted cuttings (eucalyptus) most recently genetic marker assisted selection and GMOs (pulp yield, disease resistance, etc)

Tree breeding practices breeders have had success in increasing productivity but they deal with a few species and one organisation can do all: - species trials - provenance/progeny tests - seed orchards - collection and handling - nursery production - plantation management Whereas in agroforestry we have: - a diverse client group tree species on farm - many organisations involved in the work

So tree domestication is not tree breeding! So tree domestication is not tree breeding! whilst one still undertakes trials and selection, it is also about: priority setting (species & farmers’ traits) proactive seed multiplication best nursery practices tree management extension messages (seed collection) germplasm delivery pathways marketing policies ….. and they can’t be done in isolation

Domestication defined farmer-drivenmarket-led domesticate landscapes Domesticating agroforestry trees involves accelerated and human-induced evolution to bring species into wider cultivation through a farmer-driven or market-led process. This is a science-based and iterative procedure involving the identification, production, management and adoption of high quality germplasm. High quality germplasm in agroforestry incorporates dimensions of productivity, fitness of purpose, viability and diversity. In tandem with species strategies are approaches to domesticate landscapes by investigating and modifying the uses, values, interspecific diversity, ecological functions, numbers and niches of both planted and naturally regenerated trees.

The right tree for the right place A. Trees for Products B. Trees for Services fruitfirewoodmedicineincomesawnwoodfodder soil fertility carbon sequestration soil erosion watershed protection shadebiodiversity

A farmer-driven process is the research addressing farmer’s problems? are farmers involved in the work? do farmers recognise the benefits? do farmers appreciate the benefits? are the approaches sustainable? increase production or maximise stability? are we skewing farmer’s priorities? do we understand farmer’s decision making processes?

GENERALISED TREE SPECIES DOMESTIGRAM PRIORITY SPECIES (and alternate species) Procurement (Tree Seed Suppliers Directory) Exploration Collection Conservation, Diversity Assess. IPR, Access, Exchange & Benefit Sharing EVALUATION On station On farm PROPAGATION MANAGEMENTADOPTION Laboratory On station On farm Village On stationOn farm Species trials Literature searches Expert input/meetings (historically) Farmer surveys (indigenous knowledge) On-farm/nursery tree inventories Market surveys - survival - growth - repro. ecology - genetic variation - survival - growth - niche - scale - assess. method - compatibility - indig. knowledge - molecular - nutrition - processability - seed quality - nursery practices - propagule type - symbionts - nursery practices - propagule type - indig. knowledge - spacing - watering - fertilizing - thinning - coppicing - pruning - top working - fruit set - pest/disease - reprod. ecology Collaboration & linkages - marketing/process. - outgrower schemes - germ. delivery - collective action - farmer exchanges - demonstrations - targeting climate market, soils, popn - policy - certification - incentives -environmental service payments training important for all aspects species/techniques also for domesticating the landscape PRODUCTIONIDENTIFICATION - spacing - watering - fertilizing - thinning - selection - coppicing - pruning - top working - fruit set - pest/disease - conservation - indig. Know. MULTIPLICATION - on farm - on station - Nat. Seed Cent. - composition - spacing - thinning - selection - fruit set - forecasting - timing SeedVegetative - on farm - in community - on station - Nat. Hort Cent. - no. clones - mother blocks - selection - forecasting - timing DIVERSE GERMPLASM NARROW GERMPLASM Actively Discourage (e.g. TSSD, DD, NGOs) Nomenclature (Bot. Nom. Dbase) Documentation (Agroforestree Dbase) selection

What do we know about genetic variation in tropical trees in agroforestry systems, and how do we link this to action for enhanced livelihoods and improved conservation?

Nature of the problem Farm productivity depends on both tree species diversity and genetic variation, but research on the latter has until recently not received the recognition it deserves When knowledge has become available, it has not been linked in any systematic way with management, indicating a ‘disconnect’ between research and practice

Problems in gaining information Practical and conceptual problems in gaining information on genetic variation in tree species in farm landscapes include: Lack of recognition of the nature of the problem This is related to the persistence of trees in landscapes, meaning that it can be too late to intervene by the time the problem is recognised An inability to assemble appropriate teams to undertake effective research The institutional frameworks within which researchers work rarely support the team-based approaches needed to assess variation and apply knowledge. For agroforestry, the situation is acute, as ‘forestry’ and ‘agriculture’ are traditionally considered as discrete schools of research that should be treated/taught separately

Difficulties in recognising and quantifying variation Genetic variation may be difficult to measure and important diversity may be ‘cryptic’ The large number of species involved A very large number of tree species are found in agroforestry systems, and comprehensive analysis of genetic variation in all taxa is impractical. Is the concept of ‘model’ species relevant?

Recent advances in assessing genetic variation Recent advances have been made in both direct and indirect research approaches for measuring genetic diversity in trees (These methods are described in various papers) For direct methods Morphological studies: Increased emphasis on using participatory survey techniques with communities, and on farm-forest comparisons of trees Molecular studies: Increased use of molecular markers in targeted ways that are more specific to genuine farmers’ problems and that address current concerns of the lack of practical application of these methods

For indirect methods Source surveys: Advances in methods that consult all the actors (nursery managers, local seed dealers, etc.) involved in sourcing germplasm for farmers, and through these approaches providing an indication of genetic variation in planted trees Farm inventories: Development in methods that characterise tree species found in farms and interpreting data in terms of genetic variation as well as species diversity

Current state of knowledge Based on the types of approaches to research described above, it is observed that many trees species are (These results are documented in various papers that can be provided): Subject to poor germplasm collection practice Occur at low densities in farmland Are found in highly aggregated distributions All points suggest that the effective population sizes – and therefore sustainability and productivity – of tree species in farm landscapes are on the decline, though this will depend on the functional use of species (more serious for some categories than others).

Current state of knowledge indicates that a range of germplasm-access based interventions is necessary to improve existing management practices, including: Enhancing community seed- and seedling-exchange networks, including the development of local commercial suppliers* Improving access through ‘diversity fairs’ that include both tree and crop activities (especially relevant for fruit trees) Encouraging locally-based, participatory tree domestication programmes* *ICRAF emphasis

Equally necessary, but more difficult to address, is the development of market structures that support genetic diversity in tropical tree species. Measures suggested include the development of niche markets that support a range of variation within a species (possibly using a ‘Denomination of Origin’ type approach). It is clear that tree seed and seedling supply, and product (fruit, timber, medicine, etc.) sale need to be considered as parts of one value chain if germplasm- and market-based interventions are to be successful

What resources have ICRAF and partners developed for teaching in this area? Short courses Databases Publications

Short courses The just concluded SII/World Agroforestry Centre project ‘Advancing Agroforestry Research and Development through Training and Education’, supported by The Netherlands’ Ministry of Foreign Affairs, had 20 courses for training of trainers Materials on these and other SII courses are available on CD- ROM from ICRAF’s Training Unit, and there are plans to place more of this material on-line, using the CGIAR ‘model’ for web-based learning (Jan Beniest) Notable among recent courses on the topic of tree genetic resources and domestication are:

Agroforestry and Tree Genetics: Making Markers Meaningful (2008) This course was about enabling African scientists to more effectively deploy molecular genetic markers to the field management of tree species. It was about making the linkage between technical knowledge and ground application in the context of emerging challenges to agriculture Delivering trees to farmers: improving strategies for germplasm supply (2007) This course was about bringing together the different actors involved in delivering planting material (tree seed and seedlings) to farmers, so that they can develop more productive, sustainable and environmentally-friendly agroforestry systems (this course relates to the need for germplasm-access based interventions in managing diversity, as discussed earlier) Training workshop on Allanblackia domestication (2006) This course was about developing more productive and sustainable farming systems by bringing into cultivation the Allanblackia tree, a new crop for edible oil production of interest to the global food industry. It is a case study of the tree domestication method, as a means to avoid excessive exploitation of natural resources and improve the incomes of farmers

Databases Most notable are the following: The Agroforestree Database ( provides information on more than 500 tropical trees – including timbers, fruits, fodder providers and soil fertility improving species – that are of interest for planting by smallholders. The database includes information on where species grow, how they can be propagated and managed, their uses, and pests and diseases problems (most useful of ICRAF’s online ‘tree’ databases for educational purposes) The Tree Seed Suppliers Directory ( provides information on the different suppliers of tree planting material. The database lists several thousand tree species, indicates where seed of these species can be obtained, and provides information on the quality of different seed sources. The Directory allows users to make more informed choices about the trees that they plant (more useful for field managers than for education, but useful if need to access seed for research)

Publications Tree Seeds for Farmers: A Toolkit and Reference Source (ICRAF) Describes the technical methods involved in supplying tree seed and seedlings to farmers. Describes how to go about making seed and seedling production a commercial concern (Roeland Kindt) Tree Seed Education at Agricultural and Forestry Colleges in Eastern and Southern Africa (FAO, ANAFE) Describes a possible further education curriculum on the topic (written by Christine Holding and August Temu, among others), Training in Agroforestry: A toolkit for Trainers (ICRAF) Describes the relevant methods for teaching agroforestry

Tree diversity analysis: A manual and software for common statistical methods for ecological and biodiversity studies (ICRAF) Describes how to do various statistical analyses of biodiversity data (manual and CD-ROM put together by Roeland Kindt) Molecular Markers for Tropical Trees: A Practical Guide to Principles and Procedures Description of molecular methods, protocols, and relevance for trees, (Alice Muchugi, et al) Indigenous Fruit Trees in the Tropics: Domestication, Utilization and Commercialization (ICRAF and CABI) Describes the current state of knowledge on indigenous fruit tree research