Litterbag-NIRS to Forecast Yield: a Horticultural Case with Biofertilizer Effectors

The Litterbag-NIRS method demonstrate and study the microbial biodiversity of soil in an indirect way, observing the quality decay of ground hay, under real field conditions. Three experiments pertaining to a complex microbial consortium inoculated into six horticultural species used litterbags buried for 60 days. The litter was examined, by an SCiO (TM) smart-NIR spectrometer, to extract information on the type of transformation that had taken place. Chemometric analyses of single spectra were conducted to compare any variability in three experiments. The partial least squares method was used and cross-validated to associate the observed equivalent yield indexes (YI) to the NIR spectra averaged over each productive plot, in each trial, as well as in the pooled dataset. The cross-validated R-2 values of the three experiments ranged around 0.66, and the inaccuracy of the estimates fluctuated at around +/- 5%. The pooled calibration (R-2 = 0.55) showed the presence of outlier treatments, and a marked spectral correlation (R-2 = 0.77) with the 1031 and 986 nm wavelengths. In parallel, a complex of 22 NIRS-predicted variables related to chemical decay of the hay-litter, soil characteristics, and soil microbiology was obtained and partially associated to the YI and to microbial inoculation effects. The Hay-Litterbag-NIRS method can be considered useful to indirectly demonstrate that microbial fertility is an integral part of soil fertility, as evidenced by the significant correlations and predictions of the crop yields, and by the unraveling of the tangle of plant-soil relationships.

Automated summary

The Litterbag-NIRS method indirectly demonstrates and studies microbial biodiversity in soil by observing the quality decay of ground hay. The experiments show significant correlations and predictions related to crop yields.