Measured Soil Organic Carbon Content Research Articles

Collaborative Utilization of Sentinel-1/2 and DEM Data for Mapping the Soil Organic Carbon in Forested Areas Based on the Random Forest

Optical remote sensing data are widely used for constructing soil organic carbon (SOC) mapping models. However, it is challenging to map SOC in forested areas because atmospheric water vapor affects the results derived from optical remote sensing data. To address this issue, we utilized Sentinel-1, Sentinel-2, and digital elevation model (DEM) data to obtain a comprehensive feature set (including S1-based textural indices, S2-based spectral indices, and DEM-derived indices) to map the SOC content in forested areas. The features set were the predictor variables, and the measured SOC content was the dependent variable. The random forest algorithm was used to establish the SOC model. The ratio of performance to inter-quartile range (RPIQ) was 2.92 when the S2-based spectral indices were used as predictor variables. When the comprehensive feature set was utilized as the model input, the model achieved an RPIQ of 4.13 (R2 = 0.91, root mean square error (RMSE) = 9.18), representing a 41.44% improvement in model accuracy. The average SOC content in the Greater Khingan Mountains was 43.75 g kg−1. The northern and southwestern parts had higher SOC contents (>54.93 g kg−1), while the southeastern and northwestern parts had lower contents (<39.83 g kg−1). This discrepancy was primarily attributed to agricultural activities. The results indicate that using a comprehensive feature set and the random forest algorithm is a reliable approach for estimating the spatial distribution of the SOC content in forested areas and is suitable for forest ecology and carbon management studies.

Digital Mapping of Soil Organic Carbon Using Machine Learning Algorithms in the Upper Brahmaputra Valley of Northeastern India

Soil Organic Carbon (SOC) is a crucial indicator of ecosystem health and soil quality. Machine learning (ML) models that predict soil quality based on environmental parameters are becoming more prevalent. However, studies have yet to examine how well each ML technique performs when predicting and mapping SOC, particularly at high spatial resolutions. Model predictors include topographic variables generated from SRTM DEM; vegetation and soil indices derived from Landsat satellite images predict SOC for the Lakhimpur district of the upper Brahmaputra Valley of Assam, India. Four ML models, Random Forest (RF), Cubist, Extreme Gradient Boosting (XGBoost), and Support Vector Machine (SVM), were utilized to predict SOC for the top layer of soil (0–15 cm) at a 30 m resolution. The results showed that the descriptive statistics of the calibration and validation sets were close enough to the total set data and calibration dataset, representing the complete samples. The measured SOC content varied from 0.10 to 1.85%. The RF model’s performance was optimal in the calibration and validation sets (R2c = 0.966, RMSEc = 0.159%, R2v = 0.418, RMSEv = 0.377%). The SVM model, on the other hand, had the next-lowest accuracy, explaining 47% of the variation (R2c = 0.471, RMSEc = 0.293, R2v = 0.081, RMSEv = 0.452), while the Cubist model fared the poorest in both the calibration and validation sets. The most-critical variable in the RF model for predicting SOC was elevation, followed by MAT and MRVBF. The essential variables for the Cubist model were slope, TRI, MAT, and Band4. AP and LS were the most-essential factors in the XGBoost and SVM models. The predicted OC ranged from 0.44 to 1.35%, 0.031 to 1.61%, 0.035 to 1.71%, and 0.47 to 1.36% in the RF, Cubist, XGBoost, and SVM models, respectively. Compared with different ML models, RF was optimal (high accuracy and low uncertainty) for predicting SOC in the investigated region. According to the present modeling results, SOC may be determined simply and accurately. In general, the high-resolution maps might be helpful for decision-makers, stakeholders, and applicants in sericultural management practices towards precision sericulture.

Assessing a VisNIR penetrometer system for in-situ estimation of soil organic carbon under variable soil moisture conditions

Soils are one of the main terrestrial carbon sinks and contain more carbon than the combined total in the atmosphere and terrestrial vegetation. A major challenge is quantifying how much carbon is stored in the soil and how much it has been sequestered due to changed management practices. In this study, a visible and near-infrared (VisNIR) penetrometer system was developed to measure in-situ soil VisNIR reflectance spectra to a depth of 90 cm into the soil profile. A spectral library of VisNIR soil spectra collected in air-dry and ground states were used to calibrate Cubist regression models for estimating soil organic carbon (SOC) contents. The External Parameter Orthogonalisation (EPO) transformation was applied to the in-situ spectra to remove soil moisture and other in-situ effects from the spectra. This penetrometer system was used to estimate SOC contents at three sites on three different days under variable soil moisture states. The penetrometer EPO performed better compared to the EPO constructed from moist cores. After applying EPO transformation matrix, soil wetness did not have any influence on soil SOC content estimations. The SOC contents were estimated using the in-situ VisNIR system were validated against lab-measured SOC content with mean R 2 , RMSE, and bias of 0.88, 0.32%, and 0.15%, respectively. It is concluded that the VisNIR penetrometer can potentially be used to measure SOC content rapidly and cost-effectively in-situ with a fine depth resolution. • A penetrometer system was used to acquire in-situ soil VisNIR spectra. • A spectral library from agricultural regions of NSW was used for calibration. • External Parameter Orthogonalisation (EPO) reduced confounding soil moisture effect. • External validation was performed to measure soil organic carbon contents.

Controls on vertical distribution of organic carbon in the intermontane valley soils (Barak, Northeast India)

The Intermontane valleys interspersed with rugged landscapes represent a unique ecosystem and are the barn of the eastern Himalayas (EH). Soil organic carbon (SOC) plays an important role in controlling cropland productivity in these valleys. Despite this, very little is known about SOC inventories (content and stocks) and their complex interplay with some of the major control factors such as climate, elevation, land use, parent materials, and soil properties. Therefore, the objective of this study was to assess the relative influence of heterogeneous control factors (PL: Parent materials and land use) on the vertical distribution of SOC inventories at a homogeneous climate and altitude. For the study, I choose one of the largest but largely unexplored intermontane valleys (Barak) of the eastern Himalayas. By exposing a total of 72 profiles to the C horizon or water table, a thorough analysis of the influence of PL on SOC inventories (Dichromate oxidizable) and key soil parameters controlling vertical distribution of SOC in valley soils were determined The measured SOC content in the surface soil (0.15–0.24 m deep) varied from 0.91% to 1.42%, while in the soil profile (up to a depth from 1.19 to 1.56 m), the SOC content (PWA: profile weighted average) ranged from 0.60% to 0.97%. The surface layer represented only 15% soil profile depth, while contributing 33% of the SOC content (Dichromate oxidizable). The variation in SOC content (PWA) was higher among soil parent materials (e.g. Alluvium, pebble bedded-sandstone, sandstone intercalated with shale, and pure shale, CV> 16.0%) than in land use systems (e.g. Paddy agriculture, forest, and plantation, CV <3.0%). The SOC stock on the soil profile (PWA) ranged from 19.4 to 36.0 Mg ha−1. The total (soil profile) SOC stock varied from 67.5 to 161.1 Mg ha−1. In general, the SOC inventories increased relative to the clay content (correlation, r = 0.38–0.75, p < 0.05) and parent materials controlled the variation in surface and subsurface soils while the effect of land use was mostly limited to surface soils only. Deep soils (>1.55 m) developed on shale with higher clay content had higher SOC inventories (p < 0.05) than coarse-textured sandstone soils. Parent materials contributed to greater heterogeneity (R2adj = 77.1%) in SOC inventories compared to land use systems (R2adj = 61.7%). Therefore, in addition to land use, parent materials should be considered an important control factor in the estimation of SOC inventories and future research on C- sequestration in the intermontane valleys.

National soil organic carbon map of agricultural lands in Nepal

Reliable and accurate soil organic carbon (SOC) maps are needed to monitor and improve SOC status in croplands and for agro-environmental applications. Topsoil (0–20 cm) SOC content from agricultural lands was predicted and mapped with quantified uncertainty across Nepal using state-of-the-art soil mapping techniques. Altogether 25,312 SOC observations were used to build and evaluate prediction models derived from four machine learning algorithms, namely Random Forest (RF), Cubist, Extreme Gradient Boosting (XGB) and Support Vector Machines. Twenty two environmental variables were selected as SOC predictors based on their correlation with measured SOC contents and non-collinearity with other predictors. The predictive performance of these models was compared using calibration (80% observations) and validation (20% observations) datasets. The performance of the models was also compared against a global SOC dataset compiled by International Soil Reference and Information Centre (ISRIC). The best model among the four algorithms was used to map and quantify the spatial distribution of SOC contents, and the model uncertainty was assessed with the Quantile Regression Forests technique with standard deviation representing prediction uncertainty. The RF model performed the best among all tested models, closely followed by the Cubist, and then the XGB model. The predictive performance of all of these models was better than the global SOC prediction from ISRIC. This study provides a baseline map for the topsoil SOC contents from the croplands in Nepal, and also provides a reference for similar SOC mapping studies.

Mapping Soil Organic Carbon for Airborne and Simulated EnMAP Imagery Using the LUCAS Soil Database and a Local PLSR

Soil degradation is a major threat for European soils and therefore, the European Commission recommends intensifying research on soil monitoring to capture changes over time and space. Imaging spectroscopy is a promising technique to create spatially accurate topsoil maps based on hyperspectral remote sensing data. We tested the application of a local partial least squares regression (PLSR) to airborne HySpex and simulated satellite EnMAP (Environmental Mapping and Analysis Program) data acquired in north-eastern Germany to quantify the soil organic carbon (SOC) content. The approach consists of two steps: (i) the local PLSR uses the European LUCAS (land use/cover area frame statistical survey) Soil database to quantify the SOC content for soil samples from the study site in order to avoid the need for wet chemistry analyses, and subsequently (ii) a remote sensing model is calibrated based on the local PLSR SOC results and the corresponding image spectra. This two-step approach is compared to a traditional PLSR approach using measured SOC contents from local samples. The prediction accuracy is high for the laboratory model in the first step with R2 = 0.86 and RPD = 2.77. The HySpex airborne prediction accuracy of the traditional approach is high and slightly superior to the two-step approach (traditional: R2 = 0.78, RPD = 2.19; two-step: R2 = 0.67, RPD = 1.79). Applying the two-step approach to simulated EnMAP imagery leads to a lower but still reasonable prediction accuracy (traditional: R2 = 0.77, RPD = 2.15; two-step: R2 = 0.48, RPD = 1.41). The two-step models of both sensors were applied to all bare soils of the respective images to produce SOC maps. This local PLSR approach, based on large scale soil spectral libraries, demonstrates an alternative to SOC measurements from wet chemistry of local soil samples. It could allow for repeated inexpensive SOC mapping based on satellite remote sensing data as long as spectral measurements of a few local samples are available for model calibration.

Errors induced by spectral measurement positions and instrument noise in soil organic carbon prediction using vis-NIR on intact soil

Soil spectroscopy potentially would become a routine method for the measurement of soil properties in the field on soil cores. However, many factors could lead to errors in the measurement of spectra from intact soil cores. So far, errors induced by spectral measurement positions and instrument noise have rarely been quantified. The present study evaluated these errors based on 160 intact ring core samples collected from 20 profiles in a forest, southwest China. Each ring sample was scanned using a visible near-infrared (vis-NIR) spectrometer at nine evenly distributed positions at both ends of the ring to characterize within-sample variability. In addition, 10 scans were made at each position to characterize instrument error. A position spectrum was then calculated by averaging the 10 scans at each position, while a sample spectrum was calculated by averaging all 180 scans of each sample. The samples were then tested for soil organic carbon (SOC) content using the wet oxidation method. Based on the sample spectra and SOC content, a partial least squares regression (PLSR) model was calibrated. The modeling error was empirically calculated. Each scan, position spectrum, and sample spectrum was then fed into the calibrated PLSR model to predict SOC content, and the prediction results were evaluated for errors induced by instrument noise and different spectral measurement positions. Results showed that the error induced by different spectral measurement positions was about 75% of the modeling error, while the error induced by instrument noise was small and negligible. These three errors in terms of standard error of prediction (SEP) accounted for 18.6%, 23.8% and 1.46% of the average measured SOC content of the samples, respectively. Besides, the error due to different soil core sampling positions was also considerable. In terms of SEP, this error accounted for 12.5% of the average measured SOC content. These quantified error budgets would help to account for uncertainty in the measurement of soil properties using spectroscopy in the field.

Spectroscopic measurements and imaging of soil colour for field scale estimation of soil organic carbon

Effective measurement and management of soil organic carbon (SOC) are essential for ecosystem function and food production. SOC has an important influence on soil properties and soil quality. Conventional SOC analysis is expensive and time-consuming. The development of spectral imaging sensors enables the acquisition of larger amounts of data using cheaper and faster methods. In addition, satellite remote sensing offers the potential to perform surveys more frequently and over larger areas. This research aimed to measure SOC content with colour as an indirect proxy. The measurements of soil colour were made at an agricultural site of the Czech Republic with an inexpensive digital camera and the Sentinel-2 remote sensor. Various soil colour spaces and colour indices derived from the (i) reflectance spectroscopy in the selected wavelengths of the visible (VIS) range (400–700 nm), (ii) RGB digital camera, and (iii) Sentinel-2 visible bands were used to train models for prediction of SOC. For modelling, we used the machine learning method, random forest (RF), and the models were validated with repeated 5-fold cross-validation. For prediction of SOC, the digital camera produced R2 = 0.85 and RMSEp = 0.11%, which had higher R2 and similar RMSEp compared to those obtained from the spectroscopy (R2 = 0.78 and RMSEp = 0.09%). Sentinel-2 predicted SOC with lower accuracy than other techniques; however, the results were still fair (R2 = 0.67 and RMSEp = 0.12%) and comparable with other methods. Using a digital camera with simple colour features was efficient. It enabled cheaper and accurate predictions of SOC compared to spectroscopic measurement and Sentinel-2 data.

Modeling soil organic carbon dynamics under shifting cultivation and forests using Rothc model

Shifting cultivation (jhum) and forest land are the main land uses in North Eastern region of Indian Himalaya, but in the long term this form of agriculture is not acceptable due to the soil degradation following the cutting of forest vegetation, and the consequent biodiversity loss, high erosion rates and nutrient loss through runoff. No information on soil organic carbon (SOC) dynamics and simulation studies are available, so an attempt was done using RothC model. The model was parameterized on measured SOC contents of forest and jhum sites, and average SOC stocks and changes were simulated for a period of 5 years in forest and jhum sites under the baseline and the projected climate change conditions available for Nagaland state (2021–2050). Forest sites under the baseline climate showed a steady-state condition, and simulated SOC decreased by 0.04 t C ha−1 yr−1during 5 years. In addition simulations indicated that forest land use could not benefit from climate change in a 10 years period (SOC changes showed a decreasing trend from 0.36 to 0.06 t C ha−1 yr−1). Jhum sites showed negative changes in SOC stocks both under the baseline (0.40 t C ha−1 yr−1) and the projected climate change conditions (0.50 t C ha−1 yr−1). Indeed, SOC decreases under climate change were inversely related to the duration of the jhum cropping cycle, almost linear in the first five years (from 0.65 to 0.41 t C ha−1 yr−1), and thereafter only slightly decreasing (from 0.39 to 0.32 t C ha−1 yr−1). We can conclude that under climate change conditions the jhum cropping cycle can be extended for a longer period without substantial effects on SOC decreases.

Exploring the Sensitivity of Sampling Density in Digital Mapping of Soil Organic Carbon and Its Application in Soil Sampling

The rapid monitoring and accurate estimation of dynamic changes in soil organic carbon (SOC) can make great efforts in understanding the global carbon cycle. Traditional field survey is the main approach to obtain soil data and measure SOC content. However, the limited number of soil samples and the sampling cost hinder the quality of digital soil mapping. This research aims to explore the sensitive of sampling density in digital soil mapping, and then design a suitable soil sampling plan based on a series of sampling indices. Headwall hyperspectral images (400–1700 nm) were used to estimate the SOC map by partial least squares regression (PLSR) and PLSR kriging (PLSRK). Three traditional soil sampling methods (random, grid, and Latin hypercube sampling) with 10 classes of sampling densities (6.26, 2.79, 1.57, 1.01, 0.69, 0.53, 0.39, 0.30, 0.26, and 0.20 ha−1) were designed. The R2, root mean square error (RMSE) and ratio of standard deviation to RMSE (RPD) were used to evaluate the prediction accuracy in digital soil mapping by ordinary kriging. Three new indices, namely, the ratio of sampling efficiency to performance (RSEP), the density of soil samples index and the comprehensive evaluation index of prediction accuracy, were used to select a suitable soil sampling plan. Results showed that (1) the prediction accuracy of PLSRK (RPD = 2.00) was higher by approximately 11.73% than that of PLSR (RPD = 1.79), and the hyperspectral images provided an actual referential SOC map for the study of soil sampling; (2) the grid sampling plan performed better than the random and Latin hypercube sampling methods, and the quality of SOC map improves with the increase of the sampling density, and (3) the computer simulation and field verification indicated that RSEP is one feasible index in designing a suitable soil sampling plan.

Prediction Accuracy of Soil Organic Carbon from Ground Based Visible Near-Infrared Reflectance Spectroscopy

The present study was conducted to predict soil organic carbon (SOC) from ground visible near-infrared (Vis–NIR, 400–2500 nm) spectroradiometer reflectance spectra. The objective was to study the effect of various pre-processing methods and prediction models on the accuracy of SOC estimated. Measured SOC content and reflectance spectra from pasture and cotton fields of Narrabri, Australia were used in the analysis. Reflectance spectra were pretreated with different smoothing methods such as: moving average, median filtering, gaussian smoothing and Savitzky Golay smoothing. A comparison between principal component regression, partial least square regression (PLSR) and artificial neural network models was carried out to get an optimum model for organic carbon prediction. The results indicate that PLSR model performs better with Savitzky Golay as the best pre-processing method for the study area, yielding $$ R_{\text{cal}}^{2} = \, 0.84 $$ , RPDcal = 2.55 and RPIQcal = 4.02 in the calibration set and $$ R_{\text{val}}^{2} = \, 0.77 $$ , RPDval = 2.17 and RPIQval = 3.19 in the validation set. The study recommends a suitable method in case of limited number of soil data. Based on the study, it can be said that properly pre-treated reflectance spectra show tremendous potential in soil organic carbon prediction.

Estimating carbon stocks in young moraine soils affected by erosion

In this paper the storage potential of soils within a heterogeneous structured hummocky young moraine region for organic carbon is discussed with respect to climate change and erosion. Erosion is discussed to be either a global terrestrial CO2 sink or a source. In hummocky young moraine regions of North East Germany water and tillage erosion are steadily changing factors since the beginning of arable landuse in ancient times. For such topographically complex landscapes the knowledge on soil organic carbon (SOC) dynamics and the limits of carbon storage are still limited. Our objective is to combine data collected during former soil erosion studies with recent findings on (i) soil property and (ii) estimated “optimal” SOC data to predict the SOC storage related to tillage and crop rotation, among others. Classified catenae were analysed for texture, SOC, CO3-C, nutrient contents, and depth of weathering. Optimal SOC contents were estimated on the fine sized particle content. Arable soil at convex slope positions of steep catenae show 4 time smaller SOC stocks as compared to respective forest soils and to arable soils at concave position. Our findings suggest changes in SOC stocks to be almost exclusively related to decomposable carbon pools. Comparison of estimated optimal with measured SOC contents in soils at such positions indicated that such soils could potentially store a surplus of 0.6 to 0.8gkg−1. SOC protection at convex positions is limited by soil texture, and frequent truncation of the respective soil profiles. Whereas truncation followed by downhill transfer may bury SOC at sedimentation/concave positions resulting in long-term SOC storage as far as decomposition is prevented by site conditions.

不同林龄白桦天然次生林土壤碳通量和有机碳储量

Soil respiration,known as the second largest componentof carbon dioxide flux between terrestrial ecosystems and the atmosphere,is sensitive to climate,vegetation type as well as forest age. In order to simulate the long-term forest carbon dynamics and its impact on climate system,we need to understand the successional status of carbon dynamics more than learn the responses of forest ecosystems to the changing climate. However,in spite of its significant role in determining the distribution of carbon pools and fluxes in different forest ecosystems,the effect of age on forest carbon fluxes great is uncertain. Soil temperature and soil water content are recognized as the main factors controlling the temporal variation of soil respiration. In this paper,soil carbon dynamic was measured in different age stands of Secondary Betula platyphylla( 20 a,36a,82a) in Xiaoxing'an Mountain,China. Soil respiration measurements were conducted using a LI-6400-09 soil CO2 flux system from May to October in 2008. Besides,We measured soil organic carbon content( g / kg) and bulk density( g / cm3)at four layers across the chronosequence in August 2009. Our specific objectives were to:( 1) compare the cumulative soil respiration during the growing season for the different age stands of Secondary Betula platyphylla,( 2) quantify soil organic carbon( SOC) contents and SOC density and( 3) establish quantitative relationships between soil carbon fluxes and SOC density for the three forest ecosystems. Our results showed that the seasonality of soil respiration was driven mainly by soiltemperature with the peak appeared in August,and was roughly consistent with that of soil temperature. Soil temperature at10 cm could explain 86% —92% of the variation of soil respiration for the 3 different age stands of Secondary Betula platyphylla. And the soil volumetric water content was not significantly correlated with soil respiration rate and it was not a key factor to soil respiration. We fitted site-based models and used continuous measurements of soil temperature to estimate cumulative soil respiration for the growing season of 2008( days 132—295). Cumulative soil respiration in the growing season was estimated to be 740、768 and 809 C m- 2a-1in the 20 a,36a and 82 a Betula platyphylla,respectively. Both,cumulative soil respiration and Q10 increased during stand establishment. Q10 was estimated to be 2.64,2.91 and 3.35 in the20 a,36a and 82 a Betula platyphylla,respectively. Both,total soil organic carbon( SOC) contents and SOC density at 0—50 cm increased during stand establishment,but decreased with soil depth. SOC was estimated to be 43. 75,47. 72 and55. 96 g / kg for the three forest ecosystems. Meanwhile,the figures were 14.7,18.1 and 18.7 kg / m2 for SOC density in the20 a,36a and 82 a Betula platyphylla,respectively. Soil surface carbon flux was positively correlated to SOC content( P 0. 01),and the significance level of the correlation depended on soil depth,and the soil carbon flux was more positively correlated to SOC concentration at the depth of 0—10 cm( R2= 0.908).

Simulation of SOC content and storage under different irrigation, fertilization and tillage conditions using EPIC model in the North China Plain

The objectives of the study were to use the EPIC model in simulation of soil organic carbon (SOC) content and storage under different irrigation, fertilization and tillage conditions in the North China Plain and identify the relationship of SOC change with the different field managements. There are optimal and conventional irrigation and fertilization treatments at Dongbeiwang, Quzhou and Wuqiao sites and different tillage treatments at Luancheng site. The results indicated that the simulated SOC content in Dongbeiwang, Quzhou and Wuqiao showed an increasing trend in optimal irrigation and fertilization treatments, which had the same trend with the measured data. Under different tillage conditions, the simulated SOC content in no tillage treatment reached the highest point, which was followed by rotary tillage and conventional tillage. The measured SOC content changed much compared to the simulated result in different tillage treatments. In SOC storage modeling, the optimal irrigation and fertilization treatment, no tillage treatment can sequestrate more carbon than the other treatments, which had the same trend with the measured data. The simulated highest annual change rate of SOC storage was 1.39tha−1a−1 in optimal irrigation and fertilization treatment at Wuqiao site and the lowest one was −0.18tha−1a−1 in conventional tillage treatment at Luancheng site. The SOC simulation and measurement showed optimal irrigation and fertilization, reduced tillage, and residue return were all effective strategies for improving on SOC content and storage in the North China Plain. The SOC modeling showed a good match between simulation and measurement in optimal and conventional irrigation and fertilization treatments in different sites and showed reliable trend in simulation of SOC content and storage under different irrigation, fertilization and tillage managements at different sites.

Comparing Organic Carbon Storage of Upper 15-cm Soils between Different Land Use Types in Korean Inland

This study was carried out to investigate the surface soil organic carbon fractions affecting by different land use types, including needle-leaf forest (FN), broad-leaf forest (FB), pasture, annual upland cropping land (upland), and paddy rice land (paddy). We chose seven regions across Korean inland, considering sea level altitude, and measured soil organic carbon content and physico-chemical properties such as bulk density at a depth of 0~15 cm using core samples in April for the each land use type. In addition, labile organic carbon fractions in soil including light fraction and hot water extractable carbon were investigated. From this study, organic carbon storage (Mg C per ha) in the upper 15-cm soils was highest in FB (37.8), and decreased in the order of pasture (29.1), FN (28.8), paddy (21.9), and upland crop (19.9). In forest, more than 20% of soil organic carbon existed as light fraction, the free organic matter. Hot-water extractable carbon contents of soils in five land use types were lower than 7% of their soil organic carbon content.

Removing the effect of soil moisture from NIR diffuse reflectance spectra for the prediction of soil organic carbon

Field measurement using NIR spectroscopy is becoming a popular method to provide in situ, rapid, and inexpensive estimation of soil organic carbon (SOC) content. However NIR reflectance is quite sensitive to external environmental conditions, such as temperature and soil moisture. In the field, the soil moisture content can be highly variable. It is a challenge to find a chemometric method that allows for prediction of soil organic carbon from spectra obtained under field conditions that is insensitive to variable moisture content. This paper utilises an external parameter orthogonalisation (EPO) algorithm to remove the effect of soil moisture from NIR spectra for the calibration of SOC content. The algorithm projects all the soil spectra orthogonal to the space of unwanted variation, and thus the variations of soil moisture can be effectively removed. We designed a protocol with 3 independent datasets to be used for calibration of NIR spectra: (1) the calibration dataset, which contains soil samples with measured spectra and SOC content under standard (or laboratory) condition (air-dried), (2) the EPO development dataset contains spectra under laboratory condition (air-dried samples) and spectra collected under field conditions (varying soil moisture content), and (3) the validation dataset contains spectra collected under field condition and measured SOC content. We conducted experiments using soils at different moisture contents in laboratory conditions. Using the EPO algorithm, we are able to remove the effect of soil moisture from the spectra, which resulted in improved calibration and prediction of SOC content.

Direct measurement of soil organic carbon content change in the croplands of China

Agricultural soils in China have been estimated to have a large potential for carbon sequestration, and modelling and literature survey studies have yielded contrasting results of soil organic carbon (SOC) stock change, ranging from −2.0 to +0.6% yr−1. To assess the validity of earlier estimates, we collected 1394 cropland soil profiles from all over the country and measured SOC contents in 2007–2008, and compared them with those of a previous national soil survey conducted in 1979–1982. The results showed that average SOC content in the 0–20 cm soil increased from 11.95 g kg−1 in 1979–1982 to 12.67 g kg−1 in 2007–2008, averaging 0.22% yr−1. The standard deviation of SOC contents decreased. Four major soil types had statistically significant changes in their mean SOC contents for 0–20 cm. These were: +7.5% for Anthrosols (paddy soils), +18.3% for Eutric Cambisols, +30.5% for Fluvisols, and −22.3% for Chernozems. The change of SOC contents showed a negative relationship with the average SOC contents of the two sampling campaigns only when soils in the region south of Yangtse River were excluded. SOC contents of the two major soil types in the region south of Yangtse River, i.e., Haplic Alisols/Haplic Acrisols and Anthrosols (paddy soils), changed little or significantly increased, though with a high SOC content. We suggest that the increase of SOC content is mainly attributed to the large increase in crop yields since the 1980s, and the short history as cropland establishment is mainly responsible for the decrease in SOC content for some soil types and regions showing a SOC decline.

Carbon sequestration in secondary pasture soils: a chronosequence study in the South African Highveld

Soil restoration is a means of combating desertification in semi‐arid and arid parts of the world. There, vast areas of the cropped soil degrade, particularly because of the loss of organic matter. One approach to reverse this loss is the conversion of cropland into permanent grassland for use as pasture. This study was designed to evaluate how fast and to what degree degraded cropland may re‐sequester soil organic carbon (SOC) when converted into permanent secondary pasture. Topsoil samples (0–5, 5–10 and 10–20 cm) were taken from chronosequences of secondary pastures (1 to 31 years old) at three agro‐ecosystems in the semi‐arid Highveld of South Africa. Long‐term croplands and primary grassland used as pastures served as the controls. In bulk soil samples (&lt;2 mm) and their clay (&lt;2 µm), silt (2–20 µm), fine sand (20–250 µm) and coarse sand (250–2000 µm) fractions, the contents of carbon (C) and nitrogen were determined. In all three agro‐ecosystems, using a mono‐exponential model, the SOC stocks increased exponentially until a maximum was reached 10–95 years after land conversion. This gain in SOC was clearly pronounced for the top 0–5 cm of soil, but hardly detectable at 10–20‐cm depth. The sand fractions recovered organic C more rapidly but less completely than did the finer size separates. Overall, between 9.0 and 15.3 t of SOC were sequestered in the 0–20 cm of surface soil by this land conversion. Thus, the SOC recovery in the secondary pastures resulted in SOC stocks that were 29.6–93.9% greater than those in the arable land. Yet, in no agro‐ecosystem, at any soil depth, nor in any soil fraction, did the measured SOC content reach that of the primary grassland. In part this can be attributed to a slightly finer texture of the primary grassland that had not lost silt through wind erosion or had never been used as arable land because of slightly elevated clay contents. Overall it appears, however, that previous losses of SOM cannot easily be rectified, suggesting that the native primary grassland soils are only partially resilient to land‐use change.

Detection of Carbon Stock Change in Agricultural Soils Using Spectroscopic Techniques

Soil organic carbon (SOC) represents one of the major pools in the global C cycle. Therefore, even small changes in SOC stocks cause important CO2 fluxes between terrestrial ecosystems and the atmosphere. However, SOC stocks are difficult to quantify accurately due to their high spatial variability. The aim of this paper is to evaluate the potential of Imaging Spectroscopy (IS) using the Compact Airborne Spectrographic Imager (CASI; 405–950 nm) and field spectroscopy with an Analytical Spectral Devices spectrometer (ASD; 350–2500 nm) to measure SOC content in heterogeneous agricultural soils. We used both stepwise and partial least square (PLS) regression analysis to relate spectral measurements to SOC contents. Standard Error of Prediction (SEP) for the ASD ranged from 2.4 to 3.3 g C kg−1 depending on soil moisture content of the surface layer. Imaging spectroscopy performed poorly, mainly due to the narrow spectral range of the CASI. Tests using both the CASI and the Shortwave infrared Airborne Spectrographic Imager (SASI; 900–2500 nm) showed better results. The variation in soil texture and soil moisture content degrades the spectral response to SOC contents. Currently, SEP allows to detect a SOC stock change of 7.2–9.9 Mg C ha−1 in the upper 30 cm of the soil, and is therefore still somewhat high in comparison with changes in SOC stocks as a result of management or land conversion (0.3–1.9 Mg C ha−1 yr−1). A detailed SOC maps produced by IS reflected the patterns in SOC contents due to the recent conversion from grassland to cropland.

Long-term soil organic carbon dynamics in a subhumid tropical climate: 13C data in mixed C3/C4 cropping and modeling with ROTHC

Scanty information on long-term soil organic carbon (SOC) dynamics hampers validation of SOC models in the tropics. We observed SOC content changes in a 16-year continuously cropped agroforestry experiment in Ibadan, south-western Nigeria. SOC levels declined in all treatments. The decline was most pronounced in the no-tree control treatments with continuous maize and cowpea cropping, where SOC levels dropped from the initial 15.4 to 7.3–8.0MgCha−1 in the 0–12 cm topsoil in 16 years. In the two continuously cropped alley cropping (AC) systems, one with Leucaena leucocephala and one with Senna siamea trees, SOC levels dropped to 10.7–13.2MgCha−1. Compared to the no-tree control treatments, an annual application of an additional 8.5Mgha−1 (dry matter) of plant residues, mainly tree prunings, led to an extra 3.5MgCha−1 (∼0.2% C) in the 0–12 cm top soil after 11 years, and 4.1MgCha−1 after 16 years. The addition of NPK fertilizer had little effect on the quantities of above-ground plant residues returned to the soil, and there was no evidence that the fertilizer affected the rate of SOC decomposition. The fact that both C3 and C4 plants returned organic matter to the soil in all cropping systems, but in contrasting proportions, led to clear contrasts in the 13C abundance in the SOC. This 13C information, together with the measured SOC contents, was used to test the ROTHC model. Decomposition was very fast, illustrated by the fact that we had to double all decomposition rate constants in the model in order to simulate the measured contrasts in SOC contents and δ13C between the AC treatments and the no-tree controls. We hypothesized (1) that the pruning materials from the legume trees and/or the extra rhizodeposition from the tree roots in the AC treatments accelerated the decomposition of the SOC present at the start of the experiment (true C-priming), and/or (2) that the physical protection of microbial biomass and metabolites by the clay fraction on this site, having a sandy top soil in which clay minerals are mainly of the 1:1 type, is lower than assumed by the model.