Application of Self-Organizing Maps to Explore the Interactions of Microorganisms with Soil Properties in Fruit Crops Under Different Management and Pedo-Climatic Conditions

Abstract

Background: Self-organizing maps (SOMs) are a class of neural network algorithms able to visually describe a high-dimensional dataset onto a two-dimensional grid. SOMs were explored to classify soils based on an array of physical, chemical, and biological parameters. Methods: The SOM analysis was performed considering soil physical, chemical, and microbial data gathered from an array of apple orchards and strawberry plantations managed by organic or conventional methods and located in different European climatic zones. Results: The SOM analysis considering the “climatic zone” categorical variables was able to discriminate the samples from the three zones for both crops. The zones were associated with different soil textures and chemical characteristics, and for both crops, the Continental zone was associated with microbial parameters—including biodiversity indices derived from the NGS data analysis. However, the SOM analysis based on the “management method” categorical variables was not able to discriminate the soils between organic and integrated management. Conclusions: This study allowed for the discrimination of soils of medium- and long-term fruit crops based on their pedo-climatic characteristics and associating these characteristics to some indicators of the soil biome, pointing to the possibility of better understanding the interactions among diverse variables, which could support unraveling the intricate web of relationships that define soil quality.

Keywords

• apple
• neural networks
• soil microbiome diversity
• strawberry