Soil carbon stocks, age, and respired CO2 from soybean fields and forests in Mato Grosso, Brazil

Background/Question/Methods

Primary forest and pasture are being rapidly converted to soy and maize cultivation in the southern Brazilian Amazon.  We used differences in soil carbon (C) stocks, bulk soil Δ¹⁴C and d¹³C, and  Δ¹⁴CO₂ respired during laboratory incubations to understand the changes in soil C associated with this widespread land use change. Our study area, an 800km² ranch in Mato Grosso, Brazil, was cleared for pasture land use in the early 1980s and soybean production began in the early 2000s.  We hypothesized that soil disturbance and higher temperatures in farm fields depleted C stocks relative to intact forest, and resulted in higher C turnover rates.  We tested these hypotheses by collecting soils from 0-200cm depth from four soy fields planted from 2003-2008 (7 cores per field) and seven forested sites.  We measured soil bulk density, %C, Δ¹⁴C, and δ¹³C.  We also incubated a subset (n=12 from forest and n=24 from soy) of soils in the laboratory, by wetting up dried soils to field moisture in mason jars and collecting the evolved gas.  We measured rates of CO₂ production over 4 months on a gas chromatograph (GC) and analyzed the respired Δ¹⁴CO₂ by accelerator mass spectrometry (AMS)_.

Results/Conclusions

C stocks were marginally lower (p=0.1) in the surface soils (0-10cm) of soy fields (1550±590 g/m²) than in forests (2790±260 g/m²) but did not differ for the whole soil column (forest: 22,800±6200 g/m², soy: 19,400±2450 g/m², p=0.75).  However, Δ¹⁴C of bulk soil and respired CO₂ from surface soils were lower indicating older C in soy compared to forest soils.  Forest surface soils respired CO₂ >2x faster (p<0.0001) than soy surface soils.  Respiration rates from soils at 200 cm depth did not differ between forest and soy (p=0.40) and were 18x (soy) and 67x (forest) lower than rates in surface soils.  Little change (<0.8‰) in bulk soil δ¹³C from samples collected in soy fields in 2009 and 2013 suggests a slow return towards the C3 (~ -27‰) soy signature from C4 pasture (~ -12‰), reflecting low C inputs from soy.  These data suggest: 1) conversion to pasture/soy releases relatively young C and 2) mineralization of old soil C continues for at least 30 years.  These results corroborate a pulse of soil C respiration following forest conversion to pasture/agriculture and these soils will continue to be a net CO₂ source for decades.