# Response of Soil Microbial Communities to Karst Desertification in Soil and Water Conservation Agroforestry Systems

## TL;DR

This study examines how soil microbial communities in agroforestry systems respond to different levels of karst desertification, revealing changes in soil quality and microbial diversity.

## Contribution

The study identifies a transitional optimization window in the light–moderate desertification stage for improving ecosystem function in karst agroforestry systems.

## Key findings

- Soil fertility indicators peaked at the light–moderate desertification stage, indicating optimal soil quality.
- Fungi showed stronger responses to desertification than bacteria, with Ascomycota increasing with desertification intensity.
- Soil physicochemical factors like pH, moisture, and C_N were key drivers of microbial community variation.

## Abstract

Karst desertification (KD) severely constrains regional ecological security and sustainable development. As an important ecological restoration measure, soil and water conservation agroforestry (SWCAF) systems have unclear mechanisms for soil microbial responses. This study investigated the effects of potential–light (PL), light–moderate (LM), and moderate–high (MH) KD on soil physicochemical properties and microbial communities in Karst SWCAF (KSWCAF) systems. It explored the drivers of microbial community changes. The results showed that (1) Soil physicochemical properties exhibited nonlinear changes along the KD gradient. Key soil-fertility indicators including silt, clay, total porosity (TP), total phosphorus (Total_P), total nitrogen (Total_N), soil organic carbon (SOC), and carbon nitrogen ratio (C_N) showed significant unimodal patterns, peaking at the LM stage with optimal overall soil quality; (2) The dominant bacterial phyla were Pseudomonadota, Acidobacteriota, Actinomycetota, and Planctomycetota, while the dominant fungal phyla were Ascomycota, Basidiomycota, and Mortierellomycota. The overall abundance of these dominant phyla increased with intensifying KD, except that the relative abundance of Pseudomonadota was lowest in the QZ study area, while Acidobacteriota was highest in the QZ area. The dominant fungal phylum Ascomycota increased with KD intensification; (3) KD significantly influenced microbial community structure and beta diversity. Fungi showed stronger responses to the KD gradient than bacteria. Bacterial alpha diversity was significantly higher in the LM stage compared to the PL and MH stages (p < 0.05), while fungal alpha diversity was significantly lowest in the MH stage (p < 0.05); (4) Bacterial networks exhibited highest complexity but reduced stability at the LM stage, whereas fungal networks enhanced stability at the MH stage by increasing modularization and positive correlation proportions; (5) RDA revealed that soil physicochemical factors explained 66.89% and 98.82% of bacterial and fungal community variation, respectively, with pH, moisture, and C_N as key drivers. Overall, KD regulates microbial community structure and functional allocation by reshaping the soil environmental gradient, with the LM stage potentially representing a “transitional optimization window” for KSWCAF ecosystem structure and function. This study provides a theoretical basis for microbial regulation strategies in KD control and soil and water conservation (SWC) processes.