A GLMER-based pedotransfer function expressing the relationship between total organic carbon and bulk density in forest soils

Abstract

Owing to its role in mitigating CO2 in the atmosphere, the total organic carbon (TOC) stock of soil, a key component of the terrestrial carbon cycle, is of significant interest as regards climate change. To determine TOC stock, it is first necessary to determine the soil's bulk density (BD), determined through intact soil sampling; however, in forest soils, it can be difficult to determine BD in soils with high levels of stoniness and/or tree root coverage. Furthermore, the method is time-consuming and labour-intensive, making it impractical for studies over large areas. In such cases, BD can be determined using a pedotransfer function (PTF) expressing the relationship between forest soil TOC and BD. The aim of this study was to determine a forest soil PTF using actual data obtained from 777 soil pits dug as part of the Czech Republic's National Forest Inventory (NFI). Within the NFI, BD is assessed from undisturbed core samples, while TOC is assessed from mixed samples from the same soil genetic horizons. Both generalised linear (GLM) and generalised linear mixed-effects (GLMER) models were used, with the final GLMER model best expressing the relationship for individual natural forest areas within the NFI dataset. The GLMER-based PTF described in this study can be widely applied to accurately estimate soil BD via TOC concentration at temperate forest sites where stoniness and/or root cover previously made it technically impossible to take undisturbed samples using standard methods.

Keywords

• carbon stock
• climate change
• czech national forest inventory
• czech natural forest areas
• soil properties
• soil stoniness