Resource partitioning and productivity of perennial pigeonpea/groundnut agroforestry systems in India
The productivity of two spatial arrangements of a perennial pigeonpea/groundnut agroforestry system was examined in relation to the capture and use of light and water and alterations in microclimatic conditions. Line planted (5.4 m alleys) and dispersed arrangements (1.8 x 1.2 m spacing) of pigeonpea were compared, using populations of 0.5 plants m2 for pigeonpea and 33 plants m2 (0.3 x 0.1 m spacing) for groundnut in both treatments. Sole pigeonpea and groundnut treatments were included for comparison. The experiment was conducted between July 1989 and March 1991 on a 0.6 ha plot of Alfisol at ICRISAT Center, Andhra Pradesh, India, using a randomised block design with four replications. The first groundnut harvest took place in October 1989, while pigeonpea was harvested for grain and fodder in January 1990, and was cut to a height of 0.5 m during the 1990 dry season and again in August 1990 after a second groundnut crop was sown. The second groundnut harvest took place in November 1990 and the final pigeonpea grain harvest was in January 1991. Light interception, soil and leaf temperatures and saturation deficit were continuously monitored in all treatments and at various distances from the pigeonpea in the line and dispersed treatments, whilst windspeed was monitored at a single location in each treatment. Regular destructive samples of groundnut were used to establish effects on growth and development and the results were considered in relation to the concurrent physical measurements to determine the environmental factors influencing productivity. In order to establish a water balance, rainfall records were maintained, runoff plots were installed and soil moisture content was measured regularly throughout the drying cycle. Transpiration by pigeonpea was monitored using a heat balance technique, while transpiration by groundnut and soil surface evaporation were estimated from micrometeorological data. As pigeonpea is initially slow growing, there was little reduction in groundnut yield in either the line or dispersed treatments in 1989 and there was a slight intercrop advantage in overall biomass production when expressed in the terms of the land equivalent ratios. In 1990, groundnut pod yield was reduced by 20 and 44 % in the line and dispersed treatments relative to the sole crop, despite substantial increases in the light conversion coefficient for the shaded groundnut. The lower pod yield resulted from the delayed onset of pod initiation and a slower rate of development, and was mainly due the effects of shading by the pigeonpea canopy, although mild water stress may have been a minor contributory factor. The small reductions in saturation deficit and soil and leaf temperatures experienced by the shaded groundnut had a negligible effect on growth and development. There was a considerable increase in overall biomass production in the line and dispersed treatments as compared with 1989 due to rapid pigeonpea growth, which reflected an increase in overall resource use rather than in the light conversion coefficient or water use ratios of the systems. The influence of spatial arrangement on the growth and productivity of pigeonpea became apparent after the 1990 dry season. Biomass production by pigeonpea in the dispersed treatment was approximately double that of the line planting between August 1990 and January 1991. This was entirely due to increased transpiration by the dispersed pigeonpea as a result of greater utilisation of stored soil moisture and reduced losses by surface evaporation and deep drainage. There was no difference in the water use ratio. To examine further the mechanisms responsible for the differences in productivity and water use by the line and dispersed pigeonpea, trench profile methodology was used to examine the root systems in December 1990. The root system of the dispersed pigeonpea was distributed over the entire 2.0 m depth x 2.7 m width exposed soil profile, whilst that of the line arrangement occupied no more than 50 % of the same area. The results of this work are discussed in relation to previous studies of resource use and productivity in intercropping and agroforestry systems, and possible applications and future developments are considered. Finally, the major physical and socioeconomic factors determining the potential of perennial pigeonpea/groundnut agroforestry systems for adoption by farmers in semi-arid India are discussed.