[List of Publications]

[Annual Report 97 :
Table of Contents]

Ecosystems have been defined as functional systems made up of communities of interacting organisms such as plants and animals and the environment that affects them and is in turn affected by them. Forest Ecosystem Management (FEM) requires an understanding of ecosystem function, especially the response to perturbations (e.g., removal of nutrients, genotypes, structures), its dependence on critical components (e.g., pivotal species), and the existence of thresholds and other situations where decisions may be irreversible. FEM is concerned with ecosystem function at the landscape scale; how landscape patterns and ecosystem functions interact, and how they may be modified to satisfy various objectives. Landscape scale may be interpreted as the view seen from a good vantage point on a clear day. It may approach that of a sub-catchment or small catchment, but is not necessarily delimited by watersheds.

It is unlikely that ‘business as usual’ will continue to improve the quality of life on earth. A better understanding of the environment, and better tools to predict consequences of policy and management initiatives, will enable people, planners and policy makers to enhance the quality of life. To advance the management of tropical forest ecosystems, we need to gain new knowledge, develop new techniques and promote existing knowledge. The complexity of these issues requires a holistic approach integrating social, economic and ecological aspects. An effective way to synthesise information and identify gaps in existing knowledge is to compile clear models, and test them systematically to reveal strengths, weaknesses and sensitivities.

This will highlight areas for further research as well as feasible management options. Two key elements of this work are conflict resolution and valuation of diverse goods and services.

CIFOR’s research on FEM contributes towards the better management of natural resources by offering an improved understanding of alternatives, their potential impacts and limitations and consequences of current practices. This will help to maintain ecosystem function and, in turn, lead to improved food security, conservation of biodiversity and indirectly to alleviation of poverty. A number of useful tools are being developed for those who generate policy and make decisions to accommodate the increasing pressures that societies place on forests. These techniques include the Tree Growth and Permanent Plot Information System (TROPIS) database, and an ecosystem modelling program currently under construction called FLORES, for Forest Land Oriented Resource Envisioning System.

FLORES is a model designed to assist understanding of land-use patterns in time and space, especially in forested landscapes, and to facilitate quantitative analyses of policy options to change these patterns. The program exists now in prototype. It is planned to make it available as a graphical depiction of a selected forest and all the human interventions that affect it. Jerry Vanclay, CIFOR’s systems modeller, hopes that the ultimate version of FLORES ‘may have many clients spanning natural resource managers, conservation advocates, indigenous people’s organisations, political advisors, and school children’. Potential partners in the University of Edinburgh, CIRAD, CIAT and ILRI have demonstrated interest in applying the model. The technical feasibility has been demonstrated by constructing a simple prototype with the Agroforestry Modelling Environment being developed in Edinburgh. Currently work concentrates on surveying existing data holdings to establish a suitable site for an initial calibration of the model.

The final version will resemble the popular educational computer program, SimCity, in which users construct and manage a graphical simulation of a city. Users of FLORES would make ‘what-if’ decisions involving the management of forests. As they enter different ecological and sociological variables into the model, they receive factual and visual information on the consequences of their ‘actions.’

The resulting decision-support system will allow managers and others to investigate alternatives in almost endless combinations.

TROPIS is concerned primarily with information about permanent plots and tree growth in both planted and natural forests throughout the world. During 1997, it gained increasing acceptance and a steady inflow of new contributions as well as search requests. An index of people or institutions holding plot data is maintained as well as links to valuable tools. Data can be used from comparable sites to develop models or estimates of growth for sites without empirical data.

TROPIS comprises 4 components: the TROPIS Index of Plots, the twice-a-year newsletter TROPIS-Update, the data management system MIRA and the expert system Plantgro. It now contains 12 200 plots of varying size and age, and 2900 species, and 75 search requests have been serviced. TROPIS has also been adopted as the official organ of Diversitas International – Western Pacific and Asia (DIWPA). The ‘top ten’ countries are currently Indonesia, Fiji, Brazil, Kenya, Uganda, Malaysia, Thailand, Australia, Bangladesh, and Honduras (ranked by number of plots included). Approximately 60 per cent of the plots are in natural forests, while the other 40 per cent are in plantations. The information on these plots has been contributed by 100 people in 38 countries. The plots references span the small (0.01 ha) and the large (500 ha), the new and the old (established 63 years with 17 remeasurements). A companion system for organising inventory data from plantations, known as MIRA, has gained increasing acceptance. It is now bilingual (Spanish-English), is used widely in Latin America, and is beginning to be used elsewhere (e.g., Africa).

Plant Functional Attributes (PFAs) are being currently assessed as possible indicators of site capabilities and specific growth habits. CIFOR has developed a pioneering method of characterising plants according to a set of PFAs that reflect adaptation to environment, such as leaf size class, leaf inclination and life form. New measures of plant functional diversity and complexity have been developed that complement the traditional measures based solely on species. PFAs provide a cost-efficient, uniform way of evaluating species response to land use, as well as assessing the impact of resource use on biodiversity and productivity for human needs. A strong relationship has been found between the number of unique PFA combinations (or modi) and the number of species in a given plot. In very complex plots, the existence of PFA combinations can be used to estimate plant species richness with a high degree of confidence. This technique is useful and practical for many developing country sites. Using this method, surveys across land use types in each of CIFOR’s ecoregional benchmark sites revealed unexpectedly high variation in the biodiversity of fallow systems, with some systems exhibiting levels of richness in excess of those in nearby rainforest.

CIFOR heads the above-ground biodiversity working group of the CGIAR Systemwide Programme on Alternatives to Slash and Burn which is using PFA analysis. The below-ground biodiversity group of this programme, headed by the Tropical Soil Biology and Fertility Programme, also uses functional groups for analysis. Contrary to initial expectations, data collected across a wide range of land use types showed only small effects on the number of broad functional groups. However, there were substantial shifts in species composition and density. Results from both above- and below-ground surveys will be combined to provide a holistic basis for modelling the impact of land use on biodiversity and related productivity. As study sites are developed in each of the ecoregional benchmarks, they will become increasingly important for modelling impacts of global change.

In 1997, a groundbreaking paper on PFAs entitled ‘A plant functional attribute set and grammar for dynamic vegetation description and analysis’ appeared in Functional Ecology and was well received.