Paulownia agroforestry in China - a contribution to adaptive research.

Nearly 2 million hectares of farmland have been intercropped with rows of paulownia (Paulownia elongata) trees in the North Central Plain of China. Paulownia provides an important timber source and long term income as a "saving bank". It is in harmony both ecologically and economically with the understorey winter wheat (Triticum spp.), the farmers' "survival food". Funded by ODA (Overseas Development Administration, UK), the present study investigates the role of the paulownia-crop intercropping (PCI) system in rural socio-economic development. It also identifies the socio-economic and technical factors that affect the promotion and success of PCI and examines the needs and problems encountered in the development of PCI under the present "Individual Responsibility System" (IRS). A socio-economic survey on PCI in Chengwu, Shandong Province, as a case study, shows that paulownia intercropping is an important component of socio-economic development in the poor areas. Existing policy on paulownia intercropping and land distribution causes some problems for farmers, and technical guidance on optimum tree spacing, management and maximum production of PCI with emphasis on wheat yield is eagerly awaited. An adaptive research approach to address the farmers' needs was applied in this study by interpreting scientific research to produce a useful, simple and practical model to optimise tree spacing for a range of scenarios. Data used for analysis were collected during 1983 - 1992 from an experiment station at Dangshan, Anhui Province, on a project supported by The International Development Research Centre (IDRC) of Canada. The important components of the PCI system including tree growth, growth and yield of understorey crops, and light, as a key microclimate factor affecting understorey crops, were analysed to further understanding on the natural functioning of the PCI system. The denser the spacing, the earlier the fast-growth stage of the trees occurs. However, there was no significant difference among the spacings in any of the growth variables at the end of the study in the 10th intercropping year. There are always reductions in light transmissivity (LT) and Photosynthesis Active Radiation (PAR) and the proportion of PAR to total solar radiation in the intercropped fields due to light interception by the trees. These effects are more apparent in areas near the trees, in denser spacings, in older paulownia, and when the paulownia have fully developed leaves. Intercropped wheat benefits from the modified microclimate, such as reduced temperature and increased soil water contents and relative humidity. Air temperature is the only microclimatic factor determining wheat leaf conductance, which gives a negative effect. Wheat yields start to decrease when LT is reduced to 81.8% as an average and about 10% higher for dense spacing and older trees and 10% lower for larger spacing and younger trees. Leaf conductance, biomass and yield of the intercropped wheat are slightly above those in open fields when the trees are young and widely spaced, but lower otherwise. The distribution of wheat yield across the alley in the intercropped field was similar to that of light, apart from in the dense spacing of older age trees where the competition of tree roots for nutrients probably also reduces yields near the tree rows. The reduction of leaf conductance and yield of summer crops by paulownia is greater than for the winter crop due to shading by the trees and different response of the crops. The yield reduction of cotton and maize is 15% at the edge of the tree canopy and as much as 50% near the tree rows. Regression models were developed to express DBH (used as an index of tree growth) as a function of age; light transmissivity as a function of distance across the alley for different spacings and ages of trees; understorey wheat yield as a function of distance across the alley for different spacings and ages of trees; and understorey wheat yield as a function of light transmissivity. The models can yield information on rates of tree growth and changes in understorey wheat yield and LT, by inputting tree age, and distance from west row under a given tree spacing. The output will help farmers optimise the tree-row spacing depending on whether their objective is to maximise timber production, crop production or the economic return from both. The models could also provide useful and practical technical guidance to help decision-making and settle disputes among farmers and between farmers and the local authority. Models were validated by the data collected during 1993 - 1994 from farmers' fields in Chengwu, Shandong Province. All the models perform well in these conditions and it is recommended that they be applied to PCI in all parts of North Central Plain of China due to the similarity of climate and soil conditions. Optimum tree spacings under different scenarios were also recommended. The present study demonstrated the approach of adaptive research in PCI agroforestry by linking scientific findings to rural socio-economics and farmers' needs. It indicated the equal importance of three components: socio-economic survey, experimental data analysis and interpretation, and application of scientific findings. Socio-economic survey at the village or farm level is essential for identifying the needs of farmers from PCI under IRS conditions. Data analysis helps scientists and extension officers to understand and interpret the dynamic eco-biological interactions of PCI systems . Regression modelling is a practical, efficient, simple and easy way of interpreting and applying scientific findings to a realistic PCI system.