Vertical distribution and factors influencing tropical forest soil magnetic susceptibility in Xishuangbanna, Southwest China

Introduction The measurement of soil magnetic susceptibility is rapid, nondestructive, and highly sensitive and has, therefore, been widely applied in soil research. In soil systems, the relationship between environmental factors and magnetic susceptibility is complex and interferes with the interpretation of magnetic susceptibility data. Therefore, clarifying the effects of soil factors on magnetic susceptibility and their mechanisms is necessary to explain changes in magnetic susceptibility. Magnetic characterization of tropical forest soils, which is primarily indicative of the climate, has been relatively poorly studied. Therefore, the magnetic characteristics of tropical forest soils and their correlations with soil physical and chemical properties must be systematically studied. Here, we describe a tropical rainforest soil profile that has not been disturbed by human activities and its patterns of low-frequency magnetic susceptibility (χ lf ), frequency-dependent magnetic susceptibility (χ fd %), and susceptibility of anhysteretic remanent magnetization (χ ARM ). Methods Soil sampling was conducted in six plots in Xishuangbanna, China. Soil profiles were explored in 11 layers, and various soil properties were measured. Magnetic susceptibility was assessed using susceptometry, and structural equation modeling was used to analyze the relationships between soil factors and magnetic susceptibility. Results In all profiles, the values of χ lf and χ ARM increased with depth from 0 to 30 cm, decreased with depth from 30 to 80 cm, and tended to stabilize below 80 cm. χ fd % values increased with depth from 0 to 80 cm and decreased with depth below 80 cm, with particularly rapid attenuation at the bottom of the F1 and F6 profiles. Soil properties were determined, including bulk density, water content, electrical conductivity (EC), soil particle size, soil organic carbon (SOC), pH, free iron (Fe d ), poorly crystalline iron (Fe o ), total iron (Fe t ), total nitrogen (TN), and the primary chemical elements P, S, K, Si, Al, Mn, Mg, and Zr. Linear regression and structural equation modeling were used to explore the relationships between soil factors and χ lf , to identify the main factors influencing the vertical distribution of soil χ lf, and to analyze the processes and mechanisms by which various factors affect χ lf . The results showed that the positive influence path followed EC → SOC → TN → Fe o → χ lf , and the negative influence path followed pH → N → Fe o → χ lf . Fe o was the most important factor that directly affects χ lf ( β = 0.952). The ferrihydrite in Fe o forms fine-grained magnetite and maghemite via the aging process, which is the main mechanism for increasing χ lf in the 0–80-cm layer in this area. TN had an important effect on χ lf because it affects Fe o ( β = 0.85). Discussion SOC, pH, and EC indirectly affect χ lf by promoting the conversion of ferrihydrite and silicate-bound iron to Fe o . This study highlights the role of forest soils in storing SOC and nutrients, contributing to the ecological management of forest soils.