Predictors for digital mapping of forest soil organic carbon stocks in different types of landscape

Abstract

Forest soils have a high potential to store carbon and thus mitigate climate change. The information on spatial distribution of soil organic carbon (SOC) stocks is thus very important. This study aims to analyse the importance of environmental predictors for forest SOC stock prediction at the regional and national scale in the Czech Republic. A big database of forest soil data for more than 7 000 sites was compiled from several surveys. SOC stocks were calculated from SOC content and bulk density for the topsoil mineral layer 0–30 cm. Spatial prediction models were developed separately for individual natural forest areas and for four subsets with different altitude range, using random forest method. The importance of environmental predictors in the models strongly differs between regions and altitudes. At lower altitudes, forest edaphic series and soil classes are strong predictors, while at higher altitudes the predictors related to topography become more important. The importance of soil classes depends on the pedodiversity level and on the difference in SOC stock between the classes. The contribution of forest types as predictors is limited when one (mostly coniferous) type dominates. Better prediction results can be obtained in smaller, but consistent regions, like some natural forest areas.

Digital soil mapping using machine learning-based methods to predict soil organic carbon in two different districts in the Czech Republic

“…Table 2 presents a summary of the total 16 environmental variables used in this study. Borůvka et al (2022) reported that the importance of environmental variables in the models for SOC stock prediction varies in different regions and altitudes.…”

“…In addition, changes in other soil properties (such as changes caused by cultivation) influence the SOC content prediction by affecting aggregate stability, porosity, and bulk density. It has been also found that land-use, land management, vegetation, elevation, slope, rainfall, soil type, and wetness index are the most effective predictors of SOC (Badia et al 2016;Mosleh et al 2016;Wiesmeier et al 2019;Borůvka et al 2022).…”

“…Brungard et al (2015) reported that covariates selected by soil scientists familiar with the study area did not yield the most accurate models compared to covariates automatically selected by ML algorithms. Additionally, Borůvka et al (2022) showed that even large datasets used for modelling do not guarantee highly accurate prediction. Although the number of studies using ML algorithms have been increasing, only a few studies have compared different learners, and most studies are limited to the evaluation of a few common models such as random forest (Fathololoumi et al 2020).…”

The effects of slope and altitude on soil organic carbon and clay content in different land-uses: A case study in the Czech Republic

“…Additionally, the coefficient of determination (R 2 ) between altitude and SOC was 0.144 and 0.433 for Liberec and Domažlice districts, respectively, indicating that there was a moderate correlation between altitude and SOC. Borůvka et al (2022) reported that the importance of environmental variables in the models for SOC stock prediction varies in different regions and altitudes.…”

Soil Fungi and Soil Organic Carbon Stocks in the Profile of a Forest Arenosol

“…To better understand, and potentially mitigate, future climate change, studies assessing soil organic carbon (SOC), soil fungi, and other parameters of soil are very important. Forest soils have a high potential for significant carbon sequestration [1][2][3][4]. Organic matter from aboveground and belowground vegetation is the main source of organic matter significantly affecting SOC stocks in mineral soil layers [5][6][7][8].…”