Search

• Land-use types, seasons, slope position significantly affected SOC and SWC. • Revegetation increased SWS consumption and SOC accumulation. • There was a positive linear relationship between SOC and SWC. • Planting grass and shrubs is recommended in the semi-arid hilly and gully regions. Understanding the response of soil organic carbon (SOC) and moisture to different land-use types, slope positions, seasonal variations, and their interactions has great significance for land use and management. However, the coupling effects of SOC and soil moisture in typical hilly and gully regions are still elusive. Therefore, in this study, five different land-use types, including cropland, abandoned land, grassland, shrubland, and forestland were selected in a typical Chinese loess hilly-gully region, and soil samples were collected from 0 to 200 cm profiles at the upper, middle, and lower slope positions in four seasons. The results demonstrated that the distributions of SOC and soil moisture were jointly significantly affected by land-use types, seasons, and slope positions (P < 0.01). Average SOC content was found to be highest in the shrubland, followed by grassland, abandoned land, forestland, and cropland. SOC storage at the lower slope position was 10.97% higher than that at the upper slope position. SOC storage in winter was 45.47% higher than that in summer. SOC storages of abandoned land, grassland, shrubland, and forestland were 17.38%, 33.08%, 52.52% and 1.26% more than that of cropland, respectively. Soil moisture fluctuated greatly in 0–100 cm layers and was stable beneath the 100 cm layers, which was highest in autumn, and lowest in summer. Whereas, under different land-use types, the soil moisture of cropland was the highest, especially significantly higher than all revegetated treatments in the 60–200 cm soil layers. Compared with cropland, soil water storage at other land-use types was averagely 36.45 % lower in the entire 200 cm soil profile. SOC was positively correlated with soil moisture under different factors. Thus, we recommend planting shrub and grass in watersheds with serious soil erosion in the semi-arid hilly and gully region, because they can sequestrate more SOC and retain more soil water. [ABSTRACT FROM AUTHOR]

• The mean soil organic carbon (SOC) content in agroecosystem of Eastern China was 18.24 g kg−1. • The SOC content of planted forest > cropland > grassland in the total study area. • The SOC value of grassland was higher than cropland in North China due to excessive mechanized operation. • The random forest hybrid model exhibited higher comprehensive performance in SOC prediction. • Our predictions were much more detailed and accurate than the existing SOC content maps. Soil organic carbon (SOC) plays a key role in soil function, ecosystem services, and the global carbon cycle. Large SOC stocks accumulate in agroecosystems, but the estimates of SOC distribution and magnitude in agroecosystems still have large uncertainties in global products. Based on multiple high-resolution environmental variables (terrain derived from digital elevation model, climate and organism from Landsat 8 OLI) and 369 observed soil samples from 2019 to 2021, two machine learning methods, random forest (RF) and support vector machine (SVM) models, were used to estimate the content and spatial distribution of SOC in agroecosystems of Eastern China. We proposed a hybrid model in which the optimal predictors of SOC were selected for different land use types (cropland, planted forest and grassland), and it aimed to improve the performance of machine learning. Our results showed that (1) the hybrid models performed better than the global models, and the Random forest-Hybrid (RF-Hybrid) model led to the highest prediction accuracy, with a validation R2 of 0.65 and RMSE of 5.76 g kg−1. (2) The SOC content in the agroecosystems of Eastern China among different land use types ranked in the following order: planted forest (19.92 g kg−1) > cropland (17.51 g kg−1) > grassland (17.30 g kg−1). However, in North China, the SOC content in cropland (14.04 g kg−1) was much lower than that in planted forest (17.21 g kg−1) and grassland (17.07 g kg−1), which was caused by excessive mechanized operation. (3) The agroecosystem in Northeast China presented the highest mean SOC content (26.55 g kg−1) due to low temperature. (4) Our estimations (R2 = 0.53, RMSE = 6.74 g kg−1, 30-m resolution) were more detailed and precise than the existing global SOC maps (Soil Grids: R2 = 0.26, RMSE = 12.44 g kg−1, 250-m resolution; HWSD: R2 = 0.06, RMSE = 41.07 g kg−1, 1000-m resolution). The results may improve the accuracy of agroecosystem carbon mapping and contribute to SOC assessments in agricultural ecosystems. [ABSTRACT FROM AUTHOR]

• Facility lands were considered to be greater carbon sinks compared to other land uses. • The most stable type of organic carbon is in facility land, and the most unstable is in grasslands. • Changes in land use have a significant impact on organic carbon fractions and carbon mineralization. Soil carbon is an essential component that influences soil health and agronomic productivity, and land-use practices significantly impact soil carbon stocks. In this study, the researchers investigated the impact of various land-use types (farmland, orchards, grasslands, and facility lands) on soil organic carbon (SOC) fractions in the Qaidam Basin of China's Qinghai-Tibet Plateau. The soil samples were taken from the 0–10 and 10–20 cm depths. The particulate organic carbon (POC) and mineral-associated organic carbon (MAOC) fractions of various land uses have been measured. The results revealed that grasslands had the greatest POC/SOC ratio (0.59), with POC accounting for the majority of the SOC fraction. In contrast, all other land-use types (facility land, orchards, and farmland) had greater MAOC/SOC ratios than POC/SOC. The facility land had the highest SOC content (17.75 g kg−1), and orchard had the lowest (8.51 g kg−1). The grassland had the highest POC content (9.82 g kg−1) at a depth of 0–10 cm. The trend for MAOC content among the various land-use types was facility lands > farmlands > grasslands > orchards. The soil pH was also observed to be moderately alkaline, with an average of 8. The orchard has the highest bulk density (BD), measuring 1.38 g cm−3. The grasslands contained the highest amount of total nitrogen (TN) (2.2 g kg−1) while the agricultural fields contained the highest amount of total phosphorus (TP) (1.16 g kg−1). In addition, facility lands showed the highest level of microbial activity compared to other land-use types. According to the findings, SOC, POC, and MAOC fractions had considerably positive associations with TN and microbial biomass carbon (MBC), but were significantly negatively associated with metabolic CO 2. The findings imply that different land-use types can have a considerable impact on the quality and amount of soil organic carbon, which has consequences for soil health and ecosystem processes. The findings also highlight the significance of long-term land management practices that promote the preservation and enhancement of soil organic carbon in various land-use systems. [ABSTRACT FROM AUTHOR]

• Grassland had the highest soil infiltration capacity among the land use types with vegetation. • Soil infiltration capacity of anti-dip slope was significantly higher than that of dip slope. • The bedrock strata dip could be treated as an indicator of infiltration properties. • The common experience model was the most suitable for describing the soil infiltration characteristics. The dip of bedrock has a great influence on the infiltration of water into soil, and can control the spatial distribution of rainwater in fragile karst rocky desertification regions. Thus, it is necessary to scientifically assess the soil infiltration capacity of dip and anti-dip slopes in karst trough valleys. In this study, we comprehensively evaluated the soil infiltration capacity of dip and anti-dip slopes under different land use types using the cutting ring infiltration method. The results showed the following. i) Compared with the anti-dip slope, the dip slope had a higher bulk density, noncapillary porosity and pH and a lower soil water content, capillary porosity and soil organic carbon content (p < 0.05). In addition, the soil infiltration capacity of grassland was the highest among the studied land use types except for abandoned land. ii) Both the bulk density and the total soil porosity could be treated as indicators of the infiltration properties of dip and anti-dip slopes.iii) Based on the coefficient of determination, R2 , the common experience model was the most suitable for describing the characteristics of the soil infiltration rate. iv) Based on the principal component equation, the soil infiltration capacity and organic matter of the grassland on dip slopes and anti-dip slopes were higher than those of other land use types with vegetation, but not higher than those of abandoned land without vegetation. The results of this study provide a reference for the selection of the restoration method and the assessment of the soil infiltration capacity of dip and anti-dip slopes in karst trough valleys and other karst landforms with the same geological background. [ABSTRACT FROM AUTHOR]

• Soil quality can be improved after more than sixty years of land use change from forested to grazing land. Soil quality was found to be the highest in the apple orchard area. • Cultivated and apple orchard land use increased soil quality index. • Changes in soil quality of forest and pasture areas are different. About 60 years ago, overgrazing, of severe wind erosion in Konya-Karapınar region, which is in an arid semi-arid ecosystem. In this context, an effective fight against desertification was carried out with different land management practices for erosion control. The aim of this study is to establish a minimum dataset of soil properties determine the impact of different land uses on the soil quality index (SQI). Therefore, soil properties of five different land use types [i.e.(Forested land (FL), Grazing land (GL), Cultivated land (CL), Apple orchard (AO) and Erosive area (EA))] in the wind erosion affected area were evaluated. In the study, a total of 264 disturbed and undisturbed samples were taken from three different depths (0–10, 10–20 and 20–40 cm). On these samples, Physical [field capacity (FC), wilting point (WP), bulk density, Aggregate Stability (AS), Penetration Resistance (PR)], chemical [pH, Electrical Conductivity (EC), Lime (L), Total Nitrogen (TN), Available P (AP), Ca, Mg, Na, K, exchangeable Fe, Cu, Mn, Zn Cation exchange capacity (CEC), Exchangeable sodium percentage (ESP)], and biological [Organic Carbon (OC), Active Carbon (AC), Potential Mineralizable Nitrogen (PMN), Respiration (S)] analyses were carried out. According to principal component analysis, FC, AS, ESP, CEC, AP, OC, R were selected indicators. Scoring functions of the selected indicators were created and each indicator was scored separately. Accordingly, for 0–40 cm soil depth, SQI was calculated as EA 0.10, FL 0.38, GL 0.45, CL 0.67 and AO 0.70. There have been significant improvements in soil quality compared to the EA, however, the amount of improvement has been more limited in graze and forest lands. It has been suggested that the selected indicators should be considered as the change-sensitive properties of the soils in the study area and similar areas, and that more areas should be allocated for horticulture in erosion protection areas. [ABSTRACT FROM AUTHOR]

[Display omitted] • 332 soil sample were collected from six LULC types of Eastern Indian Himalayas. • Soil physicochemical attributes perfectly classified LULC types using three ML techniques. • Soil hyperspectral data produced good LULC classification using SVM algorithm. • RF algorithm selected the influential soil and spectral variables for LULC classification. • The tested approach can be used for LULC classification of national archived soil samples. Information on periodic land use/land cover (LULC) changes are imperative for regional agricultural planning and policymaking. In this study, 332 soil samples were collected across five LULC types (coal mine degraded land, grassland, shifting cultivation or Jhum, plantation, and upland agriculture) of Eastern Indian Himalayas and subsequently characterized via traditional laboratory protocols and hyperspectral sensor. The soil physicochemical data and spectral data were separately used to classify six LULC types using three machine learning (ML) algorithms [support vector machine (SVM), random forest (RF), and K-nearest neighbors (KNN)]. Results indicated that the change in LULC directly impacted the dynamics of soil properties. The principal component analysis highlighted the interrelationships between the suite of soil physicochemical properties in classifying soil samples from different LULC types. All three ML algorithms exhibited that the physicochemical properties of the soil can perfectly classify LULC types. Spectral data from the hyperspectral sensor also demonstrated good classification accuracy. Overall, SVM performed better than RF and KNN, producing 93% and 78% classification accuracy using soil properties and spectral data, respectively. Moreover, the RF algorithm could select the influential soil and spectral variables for classifying LULC types. In the future, the approach tested herein can be used to classify several LULC types into the archived national soil spectral database. More research is needed to include a wide range of soil data and more LULC classes for a comprehensive classification using advanced deep learning tools. [ABSTRACT FROM AUTHOR]

• Cropland abandonment significantly affects soil physico-chemical properties. • Differences between revegetated practices after farmland abandonment occurred. • Afforestation triggered higher soil organic carbon than natural revegetation. • Grassland sites enhance the accumulation of soil organic carbon. • Labile fractions are higher in both vegetated sites than in grasslands. Mediterranean mountains are sensitive agroecosystems that have suffered intense land use and land cover changes (LULCC) during the last century. From the middle of the twentieth century, most of the cultivated lands in Mediterranean mountains were abandoned, allowing the recovery of vegetation (through natural revegetation and afforestation programmes). To examine the effects of farmland abandonment, secondary succession (natural revegetation) and afforestation, an intensive soil sampling was carried out in the Araguás catchment (Central Spanish Pyrenees) including sparsely vegetated areas (badlands), grasslands, shrublands and afforested sites. LULCC were mapped, and soil physico-chemical properties were analysed in reference sites (unaltered areas during the last centuries) and in the different land uses. Likewise, the soil organic carbon (SOC) content in the bulk soils and in the fractions separated by density fractionation have been studied. This study evidenced that farmland abandonment led to a mosaic landscape with different land use and land covers. Results show that LULCC significantly affect soil physico-chemical properties (soil texture, stoniness, pH, SOC, total carbon, CorgN ratio, bulk density and field capacity). Significant differences were observed between secondary and afforested sites following farmland abandonment. Afforestation triggered higher SOC than shrubland sites (natural revegetation) (1.4 and 1.1% respectively), suggesting a slower process of organic matter accumulation after farmland abandonment in the natural revegetation compared to afforestation. The significant role of grassland sites for enhancing the accumulation of SOC has been also confirmed. The results showed also significant differences in the relative contribution of each organic fraction to the bulk SOC: the amount of labile fraction (free and occluded labile fractions) is significantly higher in afforested and shrubland sites (58.1 and 51.2% respectively) than in grassland sites (36.8%). Understanding the effects of LULCC on soil properties and SOC dynamics is essential when planning post-land management practices after farmland abandonment. [ABSTRACT FROM AUTHOR]

• Spatial variation of SOC under different land uses in alpine regions was studied. • Total nitrogen was the main factor driving SOC changes in alpine soils. • Over 90% of variation in SOC was explained by the structural equation model. • Pedotransfer functions (PTFs) were used for predicting SOC content. • Our PTFs had better accuracy than published PTFs when SOC was > 40 g kg−1 Understanding soil organic carbon (SOC) distribution and influencing factors are important to accurately assess soil carbon sink potential and ecosystem quality. However, variations and controls of SOC content and stock under various land use types in the alpine region of Qinghai-Tibetan Plateau remain challenging. In this study, soil properties (pH, bulk density, porosity, clay content, soil organic carbon, soil water content, total nitrogen, total phosphorus, available nitrogen and available phosphorus) and environmental parameters (annual average temperature, annual average rainfall, normalized differential vegetation index, altitude, slope, surface roughness, longitude and latitude) were determined across a 600 km east–west transect in the Tibet alpine region for investigating the distribution characteristics and controls of SOC for three land use patterns — farmland (FA), forestland (FO) and grassland (GR). Results indicated that SOC content in the 0–30 cm decreased with increasing soil depth at different reduction rates under various land use types. The average SOC content and stock in GR was significantly higher than that in FO and FA due to the existence of mattic epipdon in GR under high altitude. The non-consideration of gravel content in soil can lead to an overestimation of SOC stock of 0–30 cm soil layer by 15.37 %, 16.70 % and 17.43 % for FA, FO and GR, respectively. Soil properties and environmental conditions simultaneously affect the spatial distribution of SOC under various land use patterns. Pedotransfer functions (PTFs) using multiple regression were used to forecast SOC content in the studied alpine region, which had a better predictive performance than previous published PTFs in the plain area, especially when SOC contents were higher than 40 g kg−1. Our study provides a foundation to build SOC prediction models for other alpine regions in the world, where SOC contents generally exhibit a wider range than the plain areas. [ABSTRACT FROM AUTHOR]

[Display omitted] • Widespread soil erosion occurred in the hilly-gully region of the Loess Plateau. • Reclamation and deforestation in 1950–1970s caused severe soil erosion. • The increased vegetation coverage since the 1990s efficiently reduced soil erosion. • Restoring forest is more appropriated measure for soil erosion mitigation. • Convex micro-topography mitigate soil loss in sloping fields. Soil erosion strongly threatens the sustainability of natural ecosystems over the world, especially in the Loess Plateau in China, where has suffered severe soil erosion along with extensive land reclamation and deforestation from the 1950s to the 1970s. The large-scale vegetation restoration practices are supposed to be effective for reducing soil erosion, but the quantitative evaluation of soil erosion, particularly the response of land utilization types and vegetation coverage changes on the mitigation of soil degradation is still not well understood. Plutonium isotopes in soil cores collected from the forest, grassland slope, apple orchard, and cornfield in a typical watershed in the hilly-gully region of the Loess Plateau were analyzed to estimate the soil erosion rates in different land-use types. Widespread soil erosion at rates of 5.1–40.5 t/ha/yr. in this region in the past six decades was estimated. The influences of human activities in the past decades, type of land utilization, level of vegetation coverage, and terrain on the soil erosion in this region were discussed, and the accumulation of the eroded soil in the study sites was explored. Different types of land utilization showed diverse soil erosion rates (forest (slight) < grassland (light) < apple orchard ≈ cornfield (moderate)), indicating that natural vegetation rehabilitation, particularly restoring forest with high vegetation coverage is a practically effective conservation measure for soil erosion control; convex micro-topography on sloping fields is critical in alleviating soil loss by depositing eroded soil. [ABSTRACT FROM AUTHOR]

• Soil organic carbon can be predicted by land use and terrain attributes. • Soil erosion can preserve SOC by burying it at downslope deposition locations. To provide a more accurate assessment of soil organic carbon (SOC) stocks in agroecosystems, sampling strategies should be designed in a way that accounts for topographic variability and soil redistribution via erosion and deposition. While the importance of topography for influencing soil properties is long established, recent advances in digital elevation models (DEMs) permit more rapid assessment of topographic attributes. We determined primary terrain attributes from a 3 m DEM to guide sampling in cultivated fields and nearby grasslands in Southeastern Minnesota, USA. Soil samples were analyzed to quantify SOC stocks and activity of 137Cs, a radioisotope employed to track soil movement over decadal timescales. In addition to soil depth, digital terrain attribute values were important for developing predictive models in both cropland and grassland sites but contributed more to predictions in cropland soils. Profile and planform curvature were significant terms in the regression model for both cropland and grassland soils. Additionally, percent slope was also significant in models for cropland soils. Overall, regression models were able to explain 90% and 73% of variability in SOC stocks observed in grassland and cropland sites, respectively. Soil profile 137Cs inventories were correlated with SOC stocks in both land cover types and show that soil movement can explain SOC variability in upland soils. Soil erosion or deposition rates (based on 137Cs inventories) ranged from 34.8 (erosion) to −70.1 (deposition) t ha−1 yr−1 in cropland sites and from 11.7 to −17.2 t ha−1 yr−1in grassland sites. Results from this study showed that soil redistribution can result in burial of eroded OC-rich surface soil in depositional locations characterized by low slopes and concave curvature in agricultural fields. [ABSTRACT FROM AUTHOR]

• Lulao catchment (LLC) is a small, homogeneous watershed on Guizhou Plateau, SW China. • Farmland soil (FS) has higher IRM 300mT /SIRM due to stronger pedogenesis in LLC. • Grassland soil (GS) has low IRM 300mT /SIRM & higher HIRM in LLC where a reservoir lies. • Contribution of FS vs GS to reservoir sediment is inferred with sediment's magnetism. • Changes in erosion is inferred as FS is distant from the reservoir than GS. This paper demonstrates that land use could enhance spatial discrimination of soil magnetism in a homogeneous watershed, and such differences help infer alterations in sediment sources and erosion. One sediment core was sampled from a reservoir built in 1958 and fifteen soil profiles from its catchment on the southern Guizhou Plateau, SW China. The whole watershed was covered by yellow earth developed on limestone. Magnetic measurements and granulometric analysis were performed on the soils and sediments while TOC and TN analysis on two soil profiles and 137Cs analysis on the sediment core. Information of seven years' land-use was derived from remote sensing images. The farmland and grassland soils had different magnetic characteristics while the downwards distributions of TOC and C/N were similar in the two profiles. The grassland soils had higher HIRM, lower χ fd % and IRM 300mT /SIRM. The higher χ fd % and IRM 300mT /SIRM of the farmland soils suggested stronger pedogenesis under cultivation. The average HIRM, χ fd % and IRM 300mT /SIRM were 38 and 334 10-6 A m2 kg−1, 5.7 and 3.6%, and 0.91 and 0.66 for the farmland and grassland soils. The farmland soils were finer probably due to tillage. The reservoir has been closely surrounded mainly by the grassland whereas the farmland, embedded with patches of the grassland, has occurred in the terrains relatively distant from the reservoir. The sediments of 1958–1985 had lower HIRM (156 10-6 A m2 kg−1) and higher IRM 300mT /SIRM (0.79), indicating greater contributions of the remote farmland soils mixed with grassland soils and stronger erosion. HIRM increased (229 10-6 A m2 kg−1) and IRM 300mT /SIRM decreased (0.67) in the sediments of 1985–2019, implying decreased distant-soil contributions and relatively increased input of the grassland soil closely around the reservoir. Erosion thus weakened, as proven by the finer sediments. The declining erosion was mainly due to farmland shrinkage and grassland expansion. [ABSTRACT FROM AUTHOR]

[Display omitted] • Advanced nonlinear algorithms were applied to predict soil CEC. • Support vector regression (SVR) and random forest (RF) were precisely targeted. • Performance of both models were tested at six land uses in different countries. • RF predicted CEC better in nearly 67% of the land uses than SVR. • SVR showed better CEC prediction with small datasets. Cation exchange capacity (CEC) is a major indicator of soil quality and nutrient retention capacity. Despite the considerable progress in CEC prediction using various models, studies to develop CEC pedotransfer functions (PTFs) using machine learning algorithms precisely, such as support vector regression (SVR) and random forest (RF), have not yet been performed in various land uses globally. This study aims to develop, evaluate, and compare the effectiveness of RF and SVR algorithms in determining CEC in different land uses that included agriculture, plantations, grasslands, forests, fallow land and deserts in five countries (Sudan, India, Italy, Iran, and Senegal). A total of 2418 soil samples were fully analyzed and clay, silt, sand, pH, and soil organic carbon (SOC) were the selected covariates for modelling. Both RF and SVR were calibrated with a training dataset (70%, 1693 samples) and validated by the remaining data (30%, 725 samples). The performance and accuracy of both models were evaluated using the Lin's concordance correlation coefficient (LCCC), root mean square error (RMSE), and normalized root mean square error (NRMSE). The accuracy of the modeling predictions was further analyzed via the Taylor diagram. The findings revealed that clay content showed a positive significant correlation with CEC in all land uses, with highest correlation in desert land use (r = 0.94; p < 0.05). Conversely, CEC was significantly and negatively correlated with sand in all land uses, with highest negative correlation obtained in desert land use (r = −0.84; p < 0.05). The RF algorithm was able to predict the CEC better than SVR in nearly 67% of the validated land use datasets precisely in desert (RMSE = 2.68 cmol c kg−1, NRMSE = 29.9%, and LCCC = 0.94), fallow land (RMSE = 5.12 cmol c kg−1, NRMSE = 55.6%, and LCCC = 0.82), forest (RMSE = 4.78 cmol c kg−1, NRMSE = 78.2%, and LCCC = 0.59), and grassland (RMSE = 8.39 cmol c kg−1, NRMSE = 50.5%, and LCCC = 0.84). Conversely, SVR better predicted CEC in agriculture (RMSE = 5.82 cmol c kg−1, NRMSE = 57.9%, and LCCC = 0.78) and plantation (RMSE = 4.64 cmol c kg−1, NRMSE = 57.9%, and LCCC = 0.74). Therefore, RF represents a promising technique to estimate soil CEC and can be used to derive effective CEC-PTFs in case of limited data availability, due to the lack of time and financial resources when the few basic soil properties are available. The findings reported in this study can be used to verify the suggested CEC-PTFs and/or their improvement. We recommend that further similar studies based on RF and SVR algorithms should consider including land use type in the Whole dataset and clay minerals in the modelling, and then compare the performance of both algorithms considering the climatic regions of the different studied countries. [ABSTRACT FROM AUTHOR]

• Pristine and agricultural top soils from a humid region (484170 km2) were surveyed. • Agriculture caused a 25–36% soil organic carbon (SOC) depletion (dSOC). • The main factor controlling dSOC was the SOC in pristine soils (SOC PRIS). • The SOC PRIS was determined by clay content, precipitation, and temperature. • Soils with a high soybean frequency in the crop rotations presented greater dSOC. The conversion of native grasslands into croplands with a high frequency of soybean (Glycine max L.) in crop rotations has diminished soil organic carbon (SOC) stocks in the Argentinean Pampas. The aims of our study were to determine the amount of SOC lost due to cultivation (dSOC) and to assess the main factors (land use, climate, and soil properties) associated with dSOC. We took paired topsoil samples (0 to 20 cm) from arable and pristine soils (n = 465) and used a path analysis approach to evaluate the direct and indirect effect of different variables on dSOC. The selected variables were SOC in pristine soils (SOC PRIS), clay content, carbon input (C input), relative soybean harvested area (Sb %), and mean annual precipitation and air temperature (PP and Temp, respectively). The percentage of sites with SOC < 20 g kg−1 was 29% in pristine soils and 66% in arable soils. The dSOC ranged from 0 to 82 Mg ha−1. Depending on the area, dSOC represented 25 to 36% of SOC PRIS stocks. The path analysis explained 60% of the dSOC variation, and the main factor controlling dSOC was SOC PRIS (by direct effect and indirect effects through clay, PP, and Temp), followed by Sb %. The Sb % depended on the productive potential of the area (soybean yield) which was associated with Temp and PP. As a conclusion, in the analysed temperate and fully humid environments, soils with greater SOC PRIS and high soybean frequency in the crop rotations presented a greater SOC depletion after conversion to agriculture. This information will be valuable when developing models to predict current and future SOC stocks. [ABSTRACT FROM AUTHOR]

• Driving factors of SOC and total N were explored in 28 lucerne fields. • The SOC and total N were mainly explained by soil resources. • Soil P, moisture, and altitude were the key factors affecting SOC and total N. • Plant species diversity positively affected SOC and total N contents. • A strong interaction effect was observed among topography, vegetation and soil on SOC. Lucerne (Medicago sativa L.) is one of the most successfully introduced species for revegetation on the Loess Plateau of China and provides important ecosystem services. However, the driving mechanism of soil organic carbon (SOC) and total nitrogen (TN) in lucerne grasslands remains unclear. This study explored the controlling factors of SOC and TN in lucerne grasslands in the semiarid Loess Plateau. A total of 112 quadrats were employed in 28 lucerne fields. Vegetation characteristics, topographic factors, and soil properties at a 0–20 cm depth were measured in each quadrat. The SOC and TN contents increased with altitude and showed positive correlations with species richness, aboveground biomass of native plants, soil moisture, soil inorganic nitrogen, total soil phosphorus (P), and C:P and N:P ratios. Variations in SOC and TN contents were mainly attributed to soil resources, followed by the interaction of topography, vegetation and soil. Soil P, soil moisture, altitude, and native plant species were the main factors controlling SOC and TN contents in these lucerne grasslands. Results suggest that specific abiotic (soil P and moisture) and biotic (plant species diversity) factors controlled SOC and TN in semiarid lucerne grasslands. These factors should be included in SOC and TN evaluation models to predict the future terrestrial ecosystem carbon and nitrogen dynamics. [ABSTRACT FROM AUTHOR]

• Macropore characteristics were quantified as a function of land use and tillage. • Land use and tillage influenced pores in surface soils (0–100 mm) substantially. • Effect of land use and tillage on pores of different sizes was not similar. • Larger soil cores can help capture variability in soil macropore characteristics. Quantification of soil macropores is important to enhance our understanding of preferential flow behavior in soils. However, knowledge of 3D soil architecture for different land uses in a continental climate region is limited. X-ray computed tomography is a novel technique that can provide valuable information about the 3D features of the soil, such as connectivity, length density, and tortuosity, that largely control water flow through the macropores. In this study, X-ray computed tomography and image analysis were employed to quantify the soil macropore characteristics from four combinations of different land uses and soil tillage practices - native grassland (Nat), alfalfa (Alf), conventional till-corn (CT), and no-till corn (NT) systems. A total of 15 undisturbed soil columns, 150 mm in diameter and 500 mm in length, were collected from different fields. The macropore characteristics were distinctively different for different land use and tillage treatments, especially near the surface depth (top 100 mm). The soil macropores under CT were highly connected and concentrated mostly in the surface soil layer. In contrast, macropores under NT were larger and had well-developed pore networks. The Alf field had relatively larger and more vertically oriented macropores as compared to the Nat field. Relatively larger macroporosity and a higher number of macropores were observed in the surface of soils collected from Nat field. Macropore characteristics were mostly similar between Alf and NT treatments. The results of this study facilitate quantitative evaluation of 3D soil macropore features with significant implications for non-equilibrium flow prediction and contaminant transport modeling in soils. [ABSTRACT FROM AUTHOR]

• Landscape-scale spatial variability of SOC in a temperate grassland was analyzed. • SOC presents strong and moderate spatial autocorrelation at two layers respectively. • Topography and land use have independent and combined influences on SOC. • Wind erosion and deposition both can alter SOC and increase its spatial variability. • The influence of wind erosion on SOC should be considered in carbon accounting. Understanding the spatial variability of soil organic carbon (SOC) and its main influencing factors is crucial for sustainable land use management and climate change mitigation. Existing studies mostly focus on the influence of environmental factors on the spatial variability of SOC, while little attention has been paid to the effects of wind erosion which has important influences in arid and semi-arid landscapes. In this study, spatial variability of SOC content at landscape scale and its main influencing factors were analysed at 0–1 cm and 1–6 cm in a typical temperate grassland. Results show that SOC content ranges from 0.45% to 4.11% and is higher at 0–1 cm. SOC presents strong and moderate spatial autocorrelation at 0–1 cm and 1–6 cm, respectively. SOC is high in part of the hilly area, and low in the west plain and southeast undulation. Elevation and vegetation coverage have significantly positive effects on SOC. SOC varies with land uses with the order that arable land (AL) < heavily grazed grassland (HG) < moderately grazed grassland (MG) < lightly grazed grassland (LG) < ungrazed grassland (UG) at 0–1 cm and AL < HG < MG < UG < LG at 1–6 cm. Under the same land use, SOC is generally the highest in the flatland at both layers and lowest in the leeward slope at 0–1 cm and the windward slope at 1–6 cm, respectively. Wind erosion and deposition can both alter SOC content and increase its spatial variability. SOC is high in areas without wind erosion and deposition and decreases with erosion and deposition rates due to the selective erosion and remote region sourced deposition materials which generally have low SOC. Such effects have important influence on land productivity, global carbon cycle and climate change, and it could not be neglected in carbon accounting. [ABSTRACT FROM AUTHOR]

• Land-use conversion was conducted to assess its effect on the changes of soil 13C. • Significant changes in δ13C were observed at the 0–10 cm depth after conversion. • SOC stock was largely increased in the 20–40 cm layer for grassland and forest. • New SOC was formed greater in grassland than in forest above 60 cm soil depth. Soil organic carbon (SOC) content and its 13C natural abundance along soil profile are sensitive to the changes of land use since these changes alter external carbon (C) inputs and belowground C turnover processes. A study was conducted to investigate how the conversion of cropland to forest, grassland or bare land (no vegetation) affect the distribution of SOC and its 13C values in 200-cm soil profiles. After 28 years of land conversion, the cropland that was remained as cropland (treated as control) had a significantly higher δ13C value than the bare land, grassland or forest at the top 10 cm soil depth. Regardless of the land use type, the mean soil 13C values in the subsurface soil layers depleted and δ13C signature decreased by 1%−5% compared with the top 10 cm soil layer, while the corresponded SOC content decreased by 3.5% to 91.2% as soil depth went down deeper. The SOC stock in the 0–200 cm profile was 16% and 12% higher in the grassland and forest, respectively, than in the control. The conservation of cropland to grassland and forest largely increased the SOC stock in the 20–40 cm soil layer compared with the control. The newly formed SOC was higher in the grassland than in the forest, and accounted for 15.4% and 9.7% of the SOC stock in the soil depth of 0–60 cm in the grassland and forest, respectively. Soil δ13C values and SOC contents were negatively correlated with pH and clay content, while they were positively correlated with nitrogen and C/N ratio along the soil profile. We infer that the conversion of cropland to forest or grassland is a beneficial practice for SOC storage, especially in the 0–60 cm depth. Such conversion has little impact on the SOC turnovers in deep soil layers during the study period. [ABSTRACT FROM AUTHOR]

• Land use systems is superior to topography in influencing soil properties. • Sites at the upper hillslope positions depict better soil qualities. • Soil properties content reflect optimal site conditions for productivity in the catchment. • The trend of soil properties is tending towards better qualities. Patterns of soil properties in highland environments remain poorly understood and limited knowledge exists on soil quality with respect to land use systems and topographic configurations. In this study, patterns of 12 surface soil physicochemical properties were assessed over a period of three years with respect to land use systems and toposequence positions in the Rwizi catchment of the Lake Victoria Basin. Four land use systems constituting mulched banana, unmulched banana, grassland and tree plantation, and three toposequence positions stratified as footslope, midslope and summit were considered and three replications were applied resulting in a total of 36 sites for soil properties measurements over a three-year period. Topographic attributes (elevation, slope gradient, slope aspect, and location) were collected at each site and correlated with surface soil properties. With the exception of pH and Ca, surface soil physicochemical properties are revealed to be optimal in the catchment. Greater spatial variation was found with soil chemical properties (8–57%) compared to physical properties (7–19%). Land use system registered significant effects (p < 0.05) for all soil chemical properties, as well as bulk density. Toposequence position is shown to have significant (p < 0.05) effects on soil pH, Soil Organic Matter, available P, N and soil texture. The effect of toposequence position revealed better soil quality along the catenary sequence summit > footslope > midslope. Strong and significant regression coefficients (r2 > 0.6, p < 0.05) for topographic attributes were found with soil pH, available P, K, Ca, Mg and sand. Overall, the land use system effect supersedes toposequence position in accounting for variations in surface soil physicochemical properties. Although environments of the Lake Victoria Basin are subject to intense agricultural activity and are usually assumed to be degraded, surface soil physicochemical properties in this part of the catchment reflect relatively healthy soil quality conditions. [ABSTRACT FROM AUTHOR]

[Display omitted] • Páramo soils are porous, humid and rich in organic matter, which buffers streamflow. • Agriculture increases macroporosity and promote loss of nutrients and organic matter. • Grazing compacts the soil and reduces porosity, infiltration and water retention capacity. • Afforestation reduces soil moisture, organic matter and water retention capacity. • Land-use change in páramos reduces streamflow buffering and increases flashiness. The páramos biome of the northern Andes is a collection of high-mountain tropical grassland wetland ecosystems that provides important ecosystem services including hydrological buffering and water supply. Human activities in these ecosystems transform vegetation cover and soil hydro-physical properties, affecting their hydrological performance and water quality and quantity. Here, we conducted a systematic review on the influence of land use (agriculture, livestock grazing, and afforestation) on the hydro-physical properties of páramo soils and analyzed its implications for streamflow buffering. Our review protocol identified 32 relevant papers, from which key hydro-physical properties linked to streamflow variability were available: soil organic matter (SOM), soil organic carbon (SOC), porosity, bulk density, saturated hydraulic conductivity, and water retention capacity (WRC). The analysis shows that soils with native cover are characterized by a porous structure that allows a high WRC and SOM content. Agriculture increases macroporosity but it leads to bare fallow plots that promote loss of nutrients and SOM. Burning generates hydrophobic aggregates that affect WRC. Livestock grazing produces soil compaction and increases bulk density, reducing infiltration and WRC. Lastly, afforestation with exotic species (e.g. pines, eucalyptus) decreases SOM and WRC by changing soil structure. In general, the analyzed land-use activities generate hydrophobic aggregates, increase bulk density, promote erosion and runoff, and impair hydrological buffering capacity. This integrated evidence from multiple empirical studies can be used to effectively communicate the effects of different land use practices on páramo soils, provide information for modelling in data-scarce situations, and contribute to decision making processes for land use planning and conservation. [ABSTRACT FROM AUTHOR]

• Soil particles had the similar parabolic relationships with soil aggregate stability. • Nitrate nitrogen plays an important role in surface soil aggregate stability. • Abandoned apple orchards can improve soil aggregate stability through a short time. Soil aggregate stability is a feasible and effective factor to understand the complex interactions between physicochemical properties and soil structure. To reveal the distributions of soil aggregate stability and its influential factors following land use change from apple orchards abandonment and development in the Nangou watershed of the Loess Plateau, China, this study selected five ages of apple orchards and their planting years were 1 year, 3 years, 6 years, 8 years and 10 years, one 15-year grassland developed from an apple orchard, one 15- year grassland developed from farmland, one natural grassland and one 15-year locust. Results showed that restored vegetation had better soil aggregate stability, soil organic carbon (SOC), and nitrogen (N) than apple orchards, and the composition of soil particles with the best aggregate stability was clay 6%, silt 8%, and sand 86%. At a 0–10 cm soil depth, soil aggregate stability had a significant positive correlation with SOC and soil total nitrogen (STN), and a negative correlation with NO 3 − and NH 4 +. In addition, vegetation diversity and coverage only affected the soil aggregate stability of the 0–10 cm soil depth; however, soil pH, bulk density, and soil aggregate-associated inorganic nitrogen were the main influential factors that drove the soil aggregate stability of the 0–30 cm soil depth. Further research discovered that macro-aggregate associated NO 3 − and micro-aggregate associated NH 4 + may be the key factors affecting the soil aggregate stability. Therefore, it is essential to further explore the effect of soil aggregate-associated inorganic nitrogen on soil aggregate stability. [ABSTRACT FROM AUTHOR]

• Multiple tracers were employed to identify the sources of eroded OM in sediments. • Gully banks were the main contributing sources to eroded bulk OM at multiple scales. • Sediment bulk OM mainly sourced from the region with light erosion intensity. • Biological production of OM impacted on the dynamics of water-extractable OM. Erosion is a fundamental phenomenon that governs the biogeochemical cycle of carbon (C) in the Earth system. Understanding the fate of eroded C at the watershed scale thus provides essential guidance for balancing the C budget and evaluating soil management strategies. In this study, we collected a total of 229 topsoils of source materials [land use types (forests, grassland, cropland, orchard, and fallow) and gully bank] and a total of 56 surface sediment samples [main river channel (18) and three tributaries (38) with different erosion intensity]. Multiple tracers (C and nitrogen (N) isotopes, molecular n-alkane biomarker, and spectroscopic indices) applied widely were then analyzed to determine relative source contributions to eroded organic matter (OM) (including bulk OM and water-extractable OM) in sediments at a broader watershed on the Loess Plateau of China. The results from an isotopic mixing model analysis showed that gully bank was the primary contributor of bulk OM in sediments, accounting for mean probability estimate (MPE) 55.3%, followed by "grassland" (MPE 21.0%), "cropland" (MPE 14.4%), and "forests" (MPE 9.3%). The contributions of grassland and cropland showed an increasing trend from upstream to downstream sections of the studied watershed. The contribution of forests was found higher in upstream than those of midstream and downstream. Along with previous evidences, our results show that erosion intensity mainly affects the proportion of C source, where gully bank was found as the highest contributor in three associated catchments, especially in region with light erosion intensity. We further demonstrated that the sediment bulk OM deposited was mainly derived from the region with light erosion intensity at the broader watershed scale. Moreover, results of n-alkane ratios and spectroscopic indices also implied that biological production of OM other than terrestrial sources (i.e., soil OM) might play a critical role in source contribution of water-extractable OM in sediments of main river channel. [ABSTRACT FROM AUTHOR]

Geological map of France showing GRSP EE contents of soils. • Land use effect of GRSP content was indirect and due to soil characteristics. • Fungal origin of GRSP could not be confirmed. • Correlation between GRSP EE and GRSP T did not support claim of contrasting ages. • Organic matter was enriched in GRSP as soils were depleted in C. Organic matter plays essential roles in soil, including physical stabilization, nutrient storage and carbon sequestration. An operationally defined fraction of soil organic matter known as glomalin or glomalin-related soil protein, GRSP, is obtained by autoclaving soil in citrate solution. It is reputed to be of fungal origin, very stable and responsible for enhanced soil physical stability. This is the first nationwide survey of GRSP content in soil and was conducted to test the hypotheses on the origin and identity of GRSP. Nearly 200 archived soils were selected on the basis of organic matter content from the French National Soil Inventory, representing mostly cropland, grassland and woodland land uses. Two fractions of GRSP were measured, easily extractable (GRSP EE) and total (GRSP T), extraction. The median values of GRSP EE and GRSP T were 0.7 and 2 g kg−1 respectively, with a strong correlation between the two fractions. Scatter plots and cubist modelling were used to explore the relationships between the contents of the two GRSP fractions and both soil properties and land use. Land use effects were almost entirely attributable to soil characteristics. No evidence was found to support the hypotheses that GRSP is solely of fungal origin, nor that easily extracted GRSP is more recent that the total fraction, although this does not disprove either hypothesis. Soil organic matter was enriched in GRSP C-depleted cropland soils and lower in C-rich woodland, this may result from inherent stability or differences in plant-related composition of carbon input. [ABSTRACT FROM AUTHOR]

• The effect of seven land-uses on an acid Mediterranean soil properties were studied. • Soil C and N content decreased with agricultural intensification. • Natural revegetation had better impact on soil properties than pine plantation. • Pine afforestation increased the soil C content but caused soil acidification. • Microbial biomass and enzymes were the best microbial indicators of soil quality. Land use change (LUC) can alter the soil quality and lead to soil degradation or soil conservation depending on the management practices. In the Mediterranean region, the impact of LUC on soil quality has been well described on neutral or alkaline soils, which are the most common in the area. However, some areas in the west of the Iberian Peninsula are covered by Raña surfaces, a continental detritic formation associated with quartzitic ranges and characterized by very acid and weathered soil. Seven following land uses were selected for this study: cork oak climax vegetation (Cork Oak), pine afforestation (Pine), natural revegetated shrubland (Shrub) and grassland (Grassland) after land abandonment, tilled olive grove (Olive), tilled annual cropland for animal forage (Tillage) and the same annual cropland managed by no-tillage (No-Tillage). The two tree-based uses, Pine and Cork oak, accumulated the highest contents of carbon (C) and nitrogen (N) fractions. However, the further acidification and the C accumulation in labile forms of Pine could hinder the feasibility of Pine plantations to restore acid and degraded soils. Instead, the natural revegetation succession after land abandonment (first Grassland and then Shrub) were found as a suitable alternative able to increase the C content and restore the microbial activity in comparison with Tillage and Olive, the two land uses which showed the lowest C and N content and reduced microbial activity. Furthermore, the improved aggregate stability in Shrub, probably caused by root exudates of dominant plant species, could play a key role in soil restoration. Microbial biomass C and N, C decomposability and enzymatic activities were the most sensible microbial indicators to discriminate among the studied land uses. A clear pattern of lower microbial biomass and enzymatic activity was found with increasing human intervention and soil disturbance. In this respect, soil managed by no-tillage contained higher soil C and N contents when compared to Tillage and the microbial indicators were comparable with those obtained under Cork Oak. Our results demonstrate that soil tillage reduction is an essential step to mitigate soil degradation in Raña soils and highlighted the suitability of No-tillage as an alternative to traditional tillage, which can increase crop productivity while building soil organic matter content. In addition, the natural soil revegetation, including natural grasslands, shrublands and climax cork oak vegetation are more preferable for degraded soil improvement when compared to pine plantations. [ABSTRACT FROM AUTHOR]

• Soil surface roughness differed greatly between different land uses and plant species. • RR was affected by near soil surface characteristics after vegetation restoration. • RR was dominated by tillage operations for land uses with tillage operations. Soil surface roughness (SSR) significantly influences hydrological and erosion processes under different conditions. Land use and its adjustment may greatly influence SSR via altering near soil surface characteristics and tillage operations, but few studies have been performed to quantify their effects on soil surface roughness. This study was conducted to investigate the variation in soil surface roughness under different land use types, and identify the main influencing factors contributing to these changes in a small watershed on the Loess Plateau, China. Eighteen sites were selected from five typical land use types (cropland, grassland, shrub land, orchard, and woodland) to measure random roughness (RR) by a photogrammetric method. The results showed that random roughness was greatly affected by land use types, plant species, and tillage operations. The mean RR under different land use types ranged from 9.47 to 10.06 mm. Woodland had the maximum RR , followed by cropland, shrub land, orchard, and grassland. For a given land use type, RR differed significantly between different plant species. The variation in random roughness was greatly influenced by the changes in near soil surface characteristics and tillage operations induced by land use adjustment. For the testing sites without tillage operations, RR increased linearly with soil median grain size, sand content, organic matter content, plant litter coverage, litter thickness, and litter density, while it decreased linearly with soil cohesion, silt content, and clay content. For the testing sites with tillage operations, RR was dominated by tillage operations. These results are helpful to understand the potential effects of vegetation restoration on hydrological and erosion processes on hillslopes in arid and semi-arid regions. [ABSTRACT FROM AUTHOR]

The natural abundance of δ15N in soils is a simple tool for understanding ecosystem nitrogen cycling. However, the main drivers for soil δ15N in different natural land use types are not clear enough. To investigate the main drivers for soil δ15N in different land uses, we measured δ15N in the top 10 cm of the soil along a precipitation gradient with various land uses on the Loess Plateau of China, which covers a total area of 620,000 km2 and is one of the most erodible area in the world, to determine effects of climatic, soil and vegetation factors on soil δ15N and the relationships between them using the partial least squares regression (PLSR) method. Soil δ15N ranged from 1‰ to 9‰ on the Loess Plateau with an average of 5.8‰ in forest lands, 5.6‰ in shrub lands, and 4.8‰ in grasslands. Calculated variable influence on projection by PLSR showed that both soil properties (including soil organic carbon content, total nitrogen content and soil texture) and mean annual precipitation influenced soil δ15N strongly. In addition, site elevation also influenced soil δ15N strongly in forest lands. The dominant controls of soil δ15N varied among three land use types on the Chinese Loess Plateau, and PLSR would be a reasonable tool to predict soil δ15N. • Soil δ15N was analyzed on the Loess Plateau and globally in different land uses. • Precipitation, soil C and N influenced soil δ15N strongly on the Loess Plateau. • Equations between soil δ15N and the main factors were built on the Loess Plateau. • Regional equations did not work on global scale due to large precipitation range. [ABSTRACT FROM AUTHOR]