Variability Of Soil Organic Carbon Content Research Articles

Integrating Machine Learning Models for Enhanced Soil Organic Carbon Estimation: A Multi-Model Fusion Approach

Machine learning approaches are utilized to identify patterns in behavior and generate predictions across various applications. The objective of this work is to create a highly efficient model for accurately measuring and analyzing the levels of soil organic carbon (SOC) in the Chambo river sub-basin, which is situated in the province of Chimborazo. The model evaluation entails the application of diverse machine learning algorithms and approaches to determine the most efficient regression model. Regression models are improved using techniques such as Artificial Neural Networks, Support Vector Machines, and Decision Trees. The Resilient Backpropagation method yields the most precise model, as it accounts for a greater proportion of the variability in SOC content for the test data. This aligns with the findings from the training data, demonstrating a relatively low mean absolute error and a processing time that is approximately 400 times faster than that of the Multilayer Perceptron algorithm. The evaluation of estimating models is an objective procedure that considers not only the findings and precise metrics derived from the model's design, but also other relevant elements. The effectiveness of the Random Forest approach, specifically the quantile regression forests technique, has been established for estimating SOC contents in the Chambo river sub-basin data.

Changes of SOC Content in China’s Shendong Coal Mining Area during 1990–2020 Investigated Using Remote Sensing Techniques

Coal mining, an important human activity, disturbs soil organic carbon (SOC) accumulation and decomposition, eventually affecting terrestrial carbon cycling and the sustainability of human society. However, changes of SOC content and their relation with influential factors in coal mining areas remained unclear. In the study, predictive models of SOC content were developed based on field sampling and Landsat images for different land-use types (grassland, forest, farmland, and bare land) of the largest coal mining area in China (i.e., Shendong). The established models were employed to estimate SOC content across the Shendong mining area during 1990–2020, followed by an investigation into the impacts of climate change and human disturbance on SOC content by a Geo-detector. Results showed that the models produced satisfactory results (R2 > 0.69, p < 0.05), demonstrating that SOC content over a large coal mining area can be effectively assessed using remote sensing techniques. Results revealed that average SOC content in the study area rose from 5.67 gC·kg−1 in 1990 to 9.23 gC·kg−1 in 2010 and then declined to 5.31 gC·Kg−1 in 2020. This could be attributed to the interaction between the disturbance of soil caused by coal mining and the improvement of eco-environment by land reclamation. Spatially, the SOC content of farmland was the highest, followed by grassland, and that of bare land was the lowest. SOC accumulation was inhibited by coal mining activities, with the effect of high-intensity mining being lower than that of moderate- and low-intensity mining activities. Land use was found to be the strongest individual influencing factor for SOC content changes, while the interaction between vegetation coverage and precipitation exerted the most significant influence on the variability of SOC content. Furthermore, the influence of mining intensity combined with precipitation was 10 times higher than that of mining intensity alone.

Effects of Urban Forest Types and Traits on Soil Organic Carbon Stock in Beijing

Forests can affect soil organic carbon (SOC) quality and distribution through forest types and traits. However, much less is known about the influence of urban forests on SOC, especially in the effects of different forest types, such as coniferous and broadleaved forests. Our objectives were to assess the effects of urban forest types on the variability of SOC content (SOC concentration (SOCC) and SOC density (SOCD)) and determine the key forest traits influencing SOC. Data from 168 urban forest plots of coniferous or broadleaved forests located in the Beijing urban area were used to predict the effects of forest types and traits on SOC in three different soil layers, 0–10 cm, 10–20 cm, and 20–30 cm. The analysis of variance and multiple comparisons were used to test the differences in SOC between forest types or layers. Partial least squares regression (PLSR) was used to explain the influence of forest traits on SOC and select the significant predictors. Our results showed that in urban forests, the SOCC and SOCD values of the coniferous forest group were both significantly higher than those of the broadleaved group. The SOCC of the surface soil was significantly higher than those of the following two deep layers. In PLSR models, 42.07% of the SOCC variance and 35.83% of the SOCD variance were explained by forest traits. Diameter at breast height was selected as the best predictor variable by comparing variable importance in projection (VIP) scores in the models. The results suggest that forest types and traits could be used as an optional approach to assess the organic carbon stock in urban forest soils. This study found substantial effects of urban forest types and traits on soil organic carbon sequestration, which provides important data support for urban forest planning and management.

Spatial variability and simulation of soil organic carbon under different land use systems: geostatistical approach

The knowledge of spatial variability of soil properties is useful for agricultural management decision making. This study was aimed at examining the spatial variability patterns of soil organic carbon (SOC) in a representative watershed in submontane Punjab, at field, landscape and watershed scales using geostatistical tools. Six land use systems from the Hoshiarpur district of Punjab, India lying in the Satluj Lower Sub basin watershed (code A01SUL11) were selected. Highest SOC content (0.26%) was found in the poplar agroforestry system and the least (0.16%) in the maize land uses system. The brick-kiln system showed a higher variability in SOC content (52%), followed by sesame (32.5%) and mango system (31.3%). Microbial biomass carbon (MBC) and dehydrogenase activity (DHA) did not show a consistently positive relation with SOC. Sensitivity analysis conducted to ascertain sample size for detecting a critical change in SOC content showed that about 5000 samples were required for detecting a 0.01% critical change in SOC in the maize land use compared to fewer samples in other systems. Kriged surface was generated by using ordinary kriging for soil organic carbon in all land use types and all variables (physical, chemical and biological) showed variable degree of spatial dependence. The study revealed the potential effects of management practices on the spatial distribution of measured parameters. Geostatistical simulation showed better performance than kriging in general. Agro-forestry systems especially poplar-based showed considerable promise in carbon sequestration. Several variables especially SOC and texture varied with change in longitude and latitude, thus signifying the need for further examination of spatial variability at watershed scale.

Constraining a coupled erosion and soil organic carbon model using hillslope‐scale patterns of carbon stocks and pool composition

AbstractA large amount of soil organic carbon (SOC) is laterally redistributed by agricultural erosion. Recent studies have shown that this leads to strong horizontal (i.e., spatial) and vertical (i.e., with soil depth) gradients in SOC stock and C pool distribution in eroding landscapes. However, the mechanisms leading to these gradients in relation to erosion and deposition are still poorly documented. In particular, the effect of the inherent properties of SOC (as controlled by the SOC pool composition) versus the effect of depth‐related soil environmental condition (i.e., differences in soil humidity, temperature, aeration, etc.) on the persistence of SOC in eroding landscapes is uncertain. Nonetheless, a detailed understanding of these factors is important to correctly assess landscape‐scale soil C turnover and vulnerability to disturbance from human activities. This study utilizes observational data on long‐term erosion/deposition rates and C pool composition derived from soil C fractionation experiments along an eroding agricultural hillslope to constrain a coupled erosion‐SOC dynamics model. The simulation results show that the data set used can result in a robust parameter estimation of a multipool C model for an eroding landscape with parameter values that are consistent with incubation experiments. A scenario analysis, where we evaluate the contribution of different processes, demonstrates that soil redistribution is essential to explain the observation that depositional locations contain more SOC in subsoils, while the SOC content of the surface layer is similar to those observed along an eroding hillslope. The spatial variability of plant production could explain some of the observed variability in SOC content, but our results suggest that the spatial variability of SOC pool composition is mainly related to soil redistribution. Finally, we suggest that environmental factors may play a more important role than the inherent properties of SOC in determining the vertical variation of SOC mineralization. This implies that depositional C stocks might be highly vulnerable to disturbance from human activities that may reconnect the buried SOC with the atmosphere.

Spatial variability of soil organic carbon in a catchment of the Loess Plateau

The issue of soil organic carbon (SOC) is of increasing concern. Because SOC, as an important soil component in farming systems, is essential for improving soil quality, sustaining food production and quality, and maintaining water quality and as a major part of the terrestrial carbon reservoir, it plays an important role in the global carbon cycle. In this paper, a total of 665 soil samples from different depths were collected randomly in the autumn of 2007, and the spatial variability of SOC content at a small catchment of the Loess Plateau was analysed using classical statistics and geo-statistical analysis. In nonsampled areas classical kriging was utilized for interpolation of SOC estimation. The classic statistical analysis revealed moderate spatial variability with all five layers of SOC-content. In addition, the average SOC content decreased with soil depth and the relationship can be modelled by an exponential equation (y=3.1795x −1.2015, R 2=0.9866) and all of the SOC-content data in the different depth were normally distributed. The geo-statistical analysis indicated a moderate spatial dependence in 0–60 cm, while in the 60–80 cm depth spatial dependence was strong. The semi-variogram could be fitted by an exponential model for 0–10 cm depth; by a spherical model for 10–20 cm depth and 60–80 cm depth; and by a Gaussian model for 20–60 cm depth. The range increases with increasing depth. In addition, classical kriging could successfully interpolate SOC content in the catchment. In general, the geo-statistics method on a watershed scale could be accurately used to evaluate spatial variability of the SOC content in the Loess Plateau, China.