Production of soybean associated with different hybrid poplar clones in a tree-based intercropping system in southwestern Québec, Canada

In tree-based intercropping systems, trees, especially hybrid poplars (HP), can compete with crops for light, water and nutrients, resulting in a decrease in crop production. Some tree management characteristics, such as plantation density and tree species selection, may help to partially control this competition. The aim of this study was to analyse the effect of HP competition on yield and yield components of a soybean intercrop. Soybean was added to a HP–hardwood plantation which was established in 2000 and composed of alternate rows spaced 6 m (field A) and 8 m (field B) apart. The experimental design permitted the comparison of the combined effects of HP clones (TD-3230, DN-3308, NM-3729,), orientation with respect to HP row (East, West) and distance from the HP row. A thinning was performed at the beginning of 2006 to increase the initial 2 m-spacing between HP within each row to a final 6 m-spacing. Percent total light transmittance (PTLT) was measured from soybean emergence to grain filling (VE-R6). Soil water content (WC) and soil N mineralisation (NMIN) were determined at two different periods corresponding to distinct soybean reproductive stages: (1) from flowering to pod formation (R1–R4) and (2) during grain filling (R5–R6). In 2005, PTLT, soil WC, soil NMIN (for R1–R4), soybean yield and yield components were significantly reduced near the HP row (2 m). The number of pods per m2 contributed more to the variation in soybean yield than did the 100-seed size. Yield and number of pods per m2 were highly correlated to PTLT, soil WC (R1–R4) and NMIN (R1–R4). An interaction between clone and orientation was found for field A. On the east side of the HP row, soybean yield with NM-3729 or DN-3308 was significantly higher than that with TD-3230, whereas it did not differ from that with the clone on the west side of the field. In field B, a significant interaction between orientation and distance was observed for PTLT and the number of pods per m2. PTLT and number of pods per m2 showed less variability on the west side compared to the east side. In 2006, a more regular PTLT distribution with respect to different orientations and distances from the HP row was observed and compared with that in 2005. Similarly, soil WC, soil NMIN, soybean yield and yield components were more uniform in 2006. These results suggest that HP clone selection, tree spacing within the rows, row spacing, orientation, and silvicultural treatments such as thinning may be useful to control the negative effects of HP competition on the intercrop.