Soybean yield determination in relay- and double-cropping in the Argentine southern pampas.

CAVIGLIA, O.P. AND SADRAS, V.O.

Foz do Iguazu - Brasil

Conferencia; VII World Soybean Research Conference. IV International Soybean Processing and Utilization Conference. III Brazilian Soybean Congress.; 2004

The most reliable option for cropping intensification  in the Argentine southern pampas is the adoption of wheat-soybean double crop (DC), in which soybean can be sown after wheat harvest (sequential, DDseq) or into standing  grain-filling wheat (relay intercropped, DCint). During two growing seasons, experiments were conducted at Balcarce (37.5º S; 58.2º W; 130 m.a.s.l.) in order to evaluate growth, yield and seed quality of double cropped soybean, both sequential and intercropped, compared with monocropped controls sown at optimum dates (SC) or at dates coincident  with those of double crops (SCint and SCseq). Early sown sole soybean crops (SC and SCint) outyielded double crops (DCseq, DCint) and SCseq in the two seasons; yield  was  linearly and positively related  to both seed mass (p<0.0001) and seed number (p<0.0001). Biomass at R7 accounted for 84-71% of variation in seed yield and harvest index accounted for 47-40%. Sole crop yields were reduced with delayed sowing at a rate of about 1.3% per day after 1 December, whereas in double crops the rate only was about 0.5%. Yields of DC ranged 74-77% of their respective control, by contrast yield of sequential cropped soybean ranged 86-110%. `Lost time to growth´, here defined as time (in days) to reach 80% of canopy full cover, was highest in relay intercropped soybean in both seasons; moreover pooling all treatments and seasons intercepted solar radiation from emergence to R1, and from emergence to R7 was negatively related to `lost time to growth´ (R2=0.60, p<0,009 and R2=0.46, p<0,04, respectively). Radiation use efficiency (RUE) computed from emergence to R7 was not significantly different among treatments. Therefore total biomass production was strongly related with radiation interception (p<0.0002) and unrelated with RUE (p>0.4). Seed yield was weakly related to crop growth rate (CGR) from R1 to R5 (p<0.01) but strongly related to CGR from R5 to R7 (p<0.006). Protein content did no differ significantly among treatments in any season whereas oil content differed only in season 2, where temperature during seed filling accounted (p<0.002) for 58% of oil content variation rather than any crop growth variable. In summary, biomass accumulation at the onset of reproductive period and temperature during seed filling were key issues related to seed yield and quality, respectively.