A Combination of Few Physicochemical and Fungal Parameters Can Explain the Soil-Dependent Variation in Wheat Biomass After Inoculation With Cladosporium tenuissimum

Abstract

Introduction

Intensive agriculture causes substantial negative impacts on agroecosystems. One approach to reduce impacts while maintaining productivity is the inoculation with beneficial microbes. Inoculants can positively affect crop growth for instance through enhancing nutrient uptake or pathogen protection. However, the efficacy of inoculants is inconsistent across different agricultural soils. In this study, we investigated to which degree the varying growth responses to an inoculant can be modelled from soil parameters.

Materials and Methods

As inoculant, we worked with the commercially available fungus Cladosporium tenuissimum and tested its effectivity on wheat plants. Variation between soils was specifically tested, while keeping other factors constant in pot experiments under controlled conditions. We assessed 25 field soils for their influence on wheat biomass response to inoculation (BRI). For each soil, we measured physicochemical parameters and characterised the soil fungal community composition. We then performed variable selection and exhaustive model screenings to find the best model explaining variations in BRI.

Results

A combined model incorporating physicochemical and fungal soil parameters outperformed models using only one of the two types of data. The best model was based on six predictors and explained 80% of the observed variability in BRI. Predictive parameters included water holding capacity and organic carbon levels as well as soil fungi of the taxa Alternaria, Cladosporium (another species than the inoculant), Acrostalagmus and Fusicolla. Organic carbon and Alternaria negatively affected the effectivity of the inoculant while the other parameters were positive predictors for inoculation success.

Conclusion

We showed that six soil parameters were sufficient to explain most of the variation of wheat responses to inoculation with C. tenuissimum. This result serves as proof-of-concept that the effectivity of inoculants can be modelled from soil parameters. It is now necessary to take this approach to practice and evaluate predictions for inoculant efficacy under field conditions.