Effects of forest management intensity on carbon and nitrogen content in different soil size fractions of a North Florida Spodosol

Abstract

Pine plantations of the southeastern USA are regional carbon (C) sinks. In spite of large increases in woody biomass due to advanced growing systems, studies have shown little or even negative eVects on the C content of the extremely sandy soils of this region. Hence, it is important to understand the mechanisms that determine the impact of intensive forest manage- ment on soil organic carbon (SOC) sequestration. This study was conducted to examine the C proWle in a 4-year-old loblolly pine (Pinus taeda L.) plantation managed under two levels of management intensity (chemical understory control and fertilizer inputs). Soil organic C and nitrogen (N) pools were evaluated using two size fractionation methods, dry and wet sieving (2000-250 m, 250-150 m, 150-53 m and <53 m). Dry sieving was preferred over wet sieving for soil size fractionation, as it preserved more structure and water- soluble SOC components such as esters and amides and did not aVect the N distribution. DiVuse ReXectance Infrared Fourier Transform Spectroscopy (DRIFTS) spectra were used to examine the chemical composition of the size fractions, which showed the presence of recently added organic matter in the largest sand frac- tion, as well as more decomposed organic matter in the <53 m fraction. Intensive forest management reduced SOC in all three 2000-53 m fractions, most likely due to reduced root input of understory plants that were con- trolled using herbicides. The 2000-250 m fractions contained nearly half of the total SOC and showed a 23% decrease in C content due to the intensive manage- ment regime. Results from this study indicated the sig- niWcance and responsiveness of sand size SOC fractions in Florida Spodosols. Results also showed that reduc- tions in SOC due to intensive management occurred after four years and highlighted the need to understand the long-term impacts and the mechanisms responsible.