Showing papers in "Catena in 2018"

PEATMAP: Refining estimates of global peatland distribution based on a meta-analysis

Abstract: PEATMAP is a GIS shapefile dataset shows a distribution of peatlands that covers the entire world. It was produced by combining the most high quality available peatland map from a wide variety of sources that describe peatland distributions at global, regional and national levels. The following sequence of comparisons to discriminate between overlapping data sources were used: (1) Relevance. The most important criterion was that source data are able to identify peatlands faithfully and to distinguish them from other land cover types, especially non-peat forming wetlands. (2) Spatial resolution. In areas where two or more overlapping data sources were indistinguishable in terms of their relevance to peatlands, the dataset with the finest spatial resolution was selected. (3) Age. In any areas where two or more overlapping datasets were indistinguishable based on both their apparent relevance to peatlands and their spatial resolution, the data product that had been most recently updated was selected. Recently updated products commonly contain much older source data, the period over which the latest revision source data were collected as the primary measure of the age of a dataset.

Landslide susceptibility mapping using J48 Decision Tree with AdaBoost, Bagging and Rotation Forest ensembles in the Guangchang area (China)

Abstract: Landslides are a manifestation of slope instability causing different kinds of damage affecting life and property. Therefore, high-performance-based landslide prediction models are useful to government institutions for developing strategies for landslide hazard prevention and mitigation. Development of data mining based algorithms shows that high-performance models can be obtained using ensemble frameworks. The primary objective of this study is to investigate and compare the use of current state-of-the-art ensemble techniques, such as AdaBoost, Bagging, and Rotation Forest, for landslide susceptibility assessment with the base classifier of J48 Decision Tree (JDT). The Guangchang district (Jiangxi province, China) was selected as the case study. Firstly, a landslide inventory map with 237 landslide locations was constructed; the landslide locations were then randomly divided into a ratio of 70/30 for the training and validating models. Secondly, fifteen landslide conditioning factors were prepared, such as slope, aspect, altitude, topographic wetness index (TWI), stream power index (SPI), sediment transport index (STI), plan curvature, profile curvature, lithology, distance to faults, distance to rivers, distance to roads, land use, normalized difference vegetation index (NDVI), and rainfall. Relief-F with the 10-fold cross-validation method was applied to quantify the predictive ability of the conditioning factors and for feature selection. Using the JDT and its three ensemble techniques, a total of four landslide susceptibility models were constructed. Finally, the overall performance of the resulting models was assessed and compared using area under the receiver operating characteristic (ROC) curve (AUC) and statistical indexes. The result showed that all landslide models have high performance (AUC > 0.8). However, the JDT with the Rotation Forest model presents the highest prediction capability (AUC = 0.855), followed by the JDT with the AdaBoost (0.850), the Bagging (0.839), and the JDT (0.814), respectively. Therefore, the result demonstrates that the JDT with Rotation Forest is the best optimized model in this study and it can be considered as a promising method for landslide susceptibility mapping in similar cases for better accuracy.

Review on landslide susceptibility mapping using support vector machines

Abstract: Landslides are natural phenomena that can cause great loss of life and damage to property. A landslide susceptibility map is a useful tool to help with land management in landslide-prone areas. A support vector machine (SVM) is a machine learning algorithm that uses a small number of samples for prediction and has been widely used in recent years. This paper presents a review of landslide susceptibility mapping using SVM. It presents the basic concept of SVM and its application in landslide susceptibility assessment and mapping. Then it compares the SVM method with four other methods (analytic hierarchy process, logistic regression, artificial neural networks and random forests) used in landslide susceptibility mapping. The application of SVM in landslide susceptibility assessment and mapping is discussed and suggestions for future research are presented. Compared with some of the methods commonly used in landslide susceptibility assessment and mapping, SVM has its strengths and weaknesses owing to its unique theoretical basis. The combination of SVM and other techniques may yield better performance in landslide susceptibility assessment and mapping. A high-quality informative database is essential and classification of landslide types prior to landslide susceptibility assessment is important to help improve model performance.

Prediction of the landslide susceptibility: Which algorithm, which precision?

Abstract: Coupling machine learning algorithms with spatial analytical techniques for landslide susceptibility modeling is a worth considering issue. So, the current research intend to present the first comprehensive comparison among the performances of ten advanced machine learning techniques (MLTs) including artificial neural networks (ANNs), boosted regression tree (BRT), classification and regression trees (CART), generalized linear model (GLM), generalized additive model (GAM), multivariate adaptive regression splines (MARS), naive Bayes (NB), quadratic discriminant analysis (QDA), random forest (RF), and support vector machines (SVM) for modeling landslide susceptibility and evaluating the importance of variables in GIS and R open source software. This study was carried out in the Ghaemshahr Region, Iran. The performance of MLTs has been evaluated using the area under ROC curve (AUC-ROC) approach. The results showed that AUC values for ten MLTs vary from 62.4 to 83.7%. It has been found that the RF (AUC = 83.7%) and BRT (AUC = 80.7%) have the best performances comparison to other MLTs.

GIS-based landslide susceptibility evaluation using a novel hybrid integration approach of bivariate statistical based random forest method

Abstract: Taibai County is a mountainous area in China, where rainfall-induced landslides occur frequently. The purpose of this study is to assess landslide susceptibility using the integrated Random Forest (RF) with bivariate Statistical Index (SI), the Certainty Factor (CF), and Index of Entropy (IOE). For this purpose, a total of 212 landslides for the study area were identified and collected. Of these landslides, 70% (148) were selected randomly for building the models and the other landslides (64) were used for validating the models. Accordingly, 12 landslide conditioning factors were considered that involve altitude, slope angle, plan curvature, profile curvature, slope aspect, distance to roads, distance to faults, distance to rivers, rainfall, NDVI, land use, and lithology. Then, the spatial correlation between conditioning factors and landslides was analysed using the RF method to quantify the predictive ability of these factors. In the next step, three landslide models, the RF-SI, RF-CF and RF-IOE, were constructed using the training dataset. Finally, the receiver operating characteristic (ROC) and statistical measures such as the kappa index, positive predictive rates, negative predictive rates, sensitivity, specificity, and accuracy were employed to validate and compare the predictive capability of the three models. Of the models, the RF-CF model has the highest positive predictive rate, specificity, accuracy, kappa index and AUC values of 0.838, 0.824, 0.865, 0.730 and 0.925 for the training data, and the highest positive predictive rate, negative predictive rate, sensitivity, specificity, accuracy, kappa index and AUC values of 0.896, 0.934, 0.938, 0.891, 0.914, 0.828, and 0.946 for the validation data, respectively. In general, the RF-CF model produced an optimized balance in terms of AUC values and statistical measures.

Effects of land use and land cover on water quality of low-order streams in Southeastern Brazil: Watershed versus riparian zone

Abstract: Land-use/land-cover (LULC) pattern influences water quality, however, this relation may be different for various spatial scales. We evaluated the LULC effects on water quality of tropical low-order streams, comparing watershed and riparian zone models. Water quality parameters were analyzed separately and together using linear mixed and multivariate models. The results indicate that the forest cover plays a significant role in keeping water clean, while agriculture and urban areas lead to water quality degradation. Pasture land had mixed effects, but in general was not correlated with poor water quality. Dissolved oxygen, phosphorus, sediment, and fecal coliforms were influenced by LULC pattern at the watershed scale, while nitrogen and organic matter were more affected by the riparian zone composition. The water quality also varies with seasonal changes in streamflow and temperature. The overall water quality variation is explained better by the LULC composition within the watershed than in the riparian zone.

Effects of vegetation and rainfall types on surface runoff and soil erosion on steep slopes on the Loess Plateau, China.

Abstract: It is widely recognized that vegetation restoration plays a key role in controlling soil erosion in China's Loess Plateau. However, the effects of vegetation types on soil erosion on steep slopes of the Loess Plateau are not yet fully understood. In this study, we carried out our experiments on surface runoff and soil loss monitoring at nine runoff plots with different vegetation types over a nine-year period from 2008 to 2016 to evaluate the effects of vegetation and rainfall on soil erosion. We classified forty-three rainfall events into three rainfall types based on a rainfall concentration index and further analyzed the sensitivities of the runoff and soil loss to these rainfall types. The results indicated that the grassland (Bothriochloa ischaemum L.) and shrubland (Sea-buckthorn) with high ground cover had a lower runoff depth and soil loss compared to the forestlands with poor ground cover with an average reduction of 50% in annual runoff depth and 92% in annual soil loss. Comparison of the mean runoff coefficient and soil loss in the three rainfall types demonstrated that rainfall events with high intensity and short duration caused more surface runoff and soil loss under all vegetation types. A power function fitted well in the runoff-soil loss relationship and the result showed that the grassland and shrubland had a smaller magnitude term which reflects less soil susceptibility to erosion. The research implies that the ground cover is an important factor in controlling soil and water loss and vegetation measures with high ground cover should be strongly recommended for soil erosion control on the Loess Plateau. It is helpful for vegetation restoration strategy and conserving soil and water on steep slopes of this area.

Tillage management impacts on soil compaction, erosion and crop yield in Stagnosols (Croatia)

Abstract: The sustainability of agroecosystems is closely related to successful soil conservation. Sustainable land use practices are crucial to reduce the impacts of agriculture on land degradation and maintain long-term soil productivity. In this context, is important to avoid practices that deteriorate the soil (e.g. soil erosion), and find the most suitable to maintain soil and crops productivity. The objective of this work is to compare the impact of different tillage systems on soil compaction, erosion and crop production on clay loam Stagnosols in Croatia. Three tillage treatments were studied: conventional tillage (CT), no-tillage (NT) and deep tillage (DT). Soil water content, bulk density and penetration resistance were determined in the 0–10, 10–20, 20–30 and 30–40 cm soil depths. Soil erosion was measured during rainfall events. The results showed that tillage treatments influenced the soil physical parameters, soil loss and crop yields. During first four years of study NT increased (p

Spatial modelling of gully erosion in Mazandaran Province, northern Iran

Abstract: Gully erosion is one of the most severe environmental problems in large areas of Iran. The spatial distribution of gully erosion and its susceptibility zonation was studied using different bivariate statistical models, such as frequency ratio (FR), weights of evidence (WofE), and index of entropy (IofE). For this purpose, 109 gully erosion locations were identified and divided into training (70%) and validating (30%) datasets. Effective factors, including elevation, slope aspect, slope degree, slope-length (LS), topographical wetness index (TWI), plan curvature, profile curvature, land use, lithology, distance from river, drainage density, and distance from road were selected to develop maps of gully erosion susceptibility. The spatial relationship between gully erosion and each effective factor was calculated by the mentioned models. The relative operating characteristic (ROC) curve was implemented for evaluating the accuracy of the applied predictive models. Results indicated that frequency ratio model had better performance (80.4%) than the weight of evidence (79.5%) and index of entropy (79%) models. The produced gully erosion susceptibility maps can be helpful to make decisions for soil and water planning and management and finally sustainable development in the Valasht watershed.

Distribution of heavy metals and metalloid in surface sediments of heavily-mined area for bauxite ore in Pengerang, Malaysia and associated risk assessment

Abstract: A detailed investigation has been conducted to evaluate the distribution of heavy metals and metalloid in the surface sediments of a bauxite mining area in association with the potential ecological and human health risk. Field sampling was carried out within the Pengerang bauxite mining areas, including mine tailings, ex-mining pond and streams. Distribution of heavy metals (Al, Cd, Co, Cr, Cu, Fe, Mn, Pb, Sr, Zn) and metalloid such as As in sediments indicated that Fe and Al constituted the greatest portion of metal elements in the sediment while Pb and Cu were found exceeding the recommended guideline values at some locations. Assessment of potential ecological risk (PERI) demonstrated low to medium ecological risk in the metal-contaminated sediments with Cd, As and Pb have generally greater risk compared to other metals, contributing the most to the total risk index (RI). The sediment enrichment factor (EF) indicated no enrichment of most metals while Pb and As at some locations were classified as having minor and moderately to severe enrichment. The geo-accumulation index (Igeo) and contamination factor (CF) indicated that the sediments were classified uncontaminated with respect to most metals. Assessment of potential human health risk revealed that the hazard index (HI) values of the carcinogenic and non-carcinogenic risks were an order of magnitude higher among children compared to adults. There were no significant non-carcinogenic risk due to metals and metalloid in the study area as HIs

Prediction of shear strength of soft soil using machine learning methods

Abstract: Shear strength of the soil is an important engineering parameter used in the design and audit of geo-technical structures. In this research, we aim to investigate and compare the performance of four machine learning methods, Particle Swarm Optimization - Adaptive Network based Fuzzy Inference System (PANFIS), Genetic Algorithm - Adaptive Network based Fuzzy Inference System (GANFIS), Support Vector Regression (SVR), and Artificial Neural Networks (ANN), for predicting the strength of soft soils. For this purpose, case studies of 188 plastic clay soil samples collected from two major projects, Nhat Tan and Cua Dai bridges in Viet Nam have been used for generating training and testing datasets for constructing and validating the models. Validation and comparison of the models have been carried out using RMSE, and R. The results show that the PANFIS has the highest prediction capability (RMSE = 0.038 and R = 0.601), followed by the GANFIS (RMSE = 0.04 and R = 0.569), SVR (RMSE = 0.044 and R = 0.549), and ANN (RMSE = 0.059 and R = 0.49). It can be concluded that out of four models the PANFIS indicates as a promising technique for prediction of the strength of soft soils.

Effect of vegetation type and season on microbial biomass carbon in Central Himalayan forest soils, India

Abstract: Soil microbial biomass is an important component of soil organic matter constituting from 2 to 5% of the soil organic carbon and play a significant role in the cycling of nutrients and overall organic matter dynamics. The present study assessed the effects of three forest types (Banj-oak forest, Chir-pine forest and Mixed oak-pine forest) on the soil physico-chemical properties and microbial biomass Carbon in Central Himalaya, India. The soil microbial biomass carbon was determined by chloroform fumigation extraction method. In the 2 year of study period, the soil microbial biomass carbon (Cmic) was significantly higher in Mixed oak-pine forest (681 ± 1.81–763 ± 1.82 μg g−1) than in the Banj-oak (518 ± 1.50–576 ± 1.73 μg g−1) and Chir-pine forest (418 ± 1.42–507 ± 2.05 μg g−1). Though insignificant, all the forest types showed distinct seasonal variations in microbial biomass carbon with a minimum value in winter season and maximum value in rainy season. The soil microbial quotients (Cmic to Corg) were higher in Chir-pine (2.52–4.18) and Banj-oak forest (2.26–4.02) than those reported in Mixed oak-pine forest (1.44–2.24). These results indicate that Mixed oak-pine forest is better in sustaining the soil microbial biomass and soil nutrients than Banj-oak and Chir-pine forest. It recommends that nutrients rich Mixed oak-pine forest should be preferred as a forest management practice to promote microbial diversity, their activities and soil quality enhancement in Central Himalayan forests.

Ecological stoichiometry in leaves, roots, litters and soil among different plant communities in a desertified region of Northern China

Abstract: Ecological stoichiometry reflects the element content and energy flow, which are important for biogeochemical cycling in ecosystems. However, the ecological stoichiometry in leaves, roots, litter and soil is largely unknown, especially in the desertified region of Northern China. Here, six dominant plant communities (Stipa bungeana, Agropyron mongolicum, Glycyrrhiza uralensis, Cynanchum komarovii, Artemisia ordosica, and Sophora alopecuroides) were collected, and the carbon (C), nitrogen (N) and phosphorus (P) contents of leaves, roots, litters and soil were measured to explore the C:N:P stoichiometry and its driving factors. The C:N:P stoichiometry in leaves, roots, litters, and soil varied widely, and the plant community had a significant effect on the C:N:P stoichiometry in this region. There were high soil C:N, C:P and N:P ratios in non-leguminous plant communities and a high leaf N:P ratio in leguminous plant communities, and the C:N and C:P ratios in leaves were higher than in those in roots in all plant communities (p

Distribution, source and pollution level of heavy metals in river sediments from South China

Abstract: The sediment pollution caused by heavy metals has attracted a great deal of attention due to its persistence, bioaccumulation and toxicity. This research was the first to consider the whole of South China to obtain an overall profile of heavy metal spatial distribution, possible sources and pollution levels in river systems. For these data, 14 selected heavy metals were analysed in river sediments collected from sampling sites in Guangdong, Fujian, Guangxi and Hainan Provinces. The geoaccumulation index and enrichment factor revealed that river systems in South China were universally contaminated by Cd, As and Sn, which might be distributed by anthropogenic activities. Moreover, Guangdong Province, a relatively developed area in South China, was relatively polluted by certain heavy metals such as Ni, Cu, Zn and Mn. Multivariate statistical analyses such as Pearson's correlation matrix and a principal component analysis determined that several of the heavy metals might be derived from similar anthropogenic activities such as industrial effluents and domestic sewage discharge. In terms of heavy metal contamination in South China, necessary measures should be undertaken to protect rivers in South China.

Responses of vegetation productivity to multi-scale drought in Loess Plateau, China

Abstract: Drought affects land surface dynamics. Quantifying the response of vegetation productivity to variations in drought events at different time-scales is crucial for evaluating the potential impacts of climate change on terrestrial ecosystems. Utilizing the Standardized Precipitation Evapotranspiration Index (SPEI) and Normalized Difference Vegetation Index (NDVI), this study evaluated the response of vegetation productivity to different time-scales of drought (SPEI-3, SPEI-6, SPEI-12, and SPEI-24, with 3, 6, 12 and 24 months of accumulation, respectively) in the growing season (April to October), as well as the spring, summer and autumn of the Loess Plateau (LP) by the maximum Pearson correlation (rmax). Results indicated that: (1) major areas (91.49%, 88.81%, 94.41% and 79.20%) of the LP were highly controlled by drought at the different time-scales during 1982–2013. However, high spatial and seasonal differences occurred during different time-scales, with the maximum influence in summer at 3-month time (SPEI-3); (2) rmax showed that 98.47%, 45.91%, 89.80% and 75.33% of the LP show significant correlation (P

Effects of topographic factors on runoff and soil loss in Southwest China

Abstract: Soil erosion is a threat to sustainable agricultural and regional development in karst regions. In this study, field plot observation method was used to estimate the effects of slope gradient and length on runoff and soil loss in Guizhou, Southwest China. The results showed that runoff and soil loss is nonlinearly related to slope gradient. The increasing trends of runoff and soil loss declined after the slope gradient of 15°. This turning point was affected by both slope gradient and rock outcrops on the 20°–25° slopes, hence it is still unknown whether the slope gradient of 15° is a critical value. Runoff showed a trend of decrease-increase-decrease as slope length increased, and soil loss rate showed an increasing trend as slope length increased. There is a significantly positive linear relationship between soil loss and slope length (P < 0.01). Runoff and soil loss were significantly correlated with rainfall amount (P) and the maximum 30 or 60 min rainfall intensity (I30 or I60), which had power function with PI30 on gradient-changed slopes and PI60 on length-changed slopes. Moreover, soil loss has a power function relationship with slope gradient/length and runoff depth. This study is helpful to elucidate the effect of topographic factors on soil erosion and to take effective soil conservation measures in karst regions.

The effect of rice husk biochar on soil nutrient status, microbial biomass and paddy productivity of nutrient poor agriculture soils

Abstract: The study related to the effects of rice husk biochar (RHB) application on soil microbial aspects and paddy productivity in field condition is scare. Therefore, present study provides fresh insight into the effects of RHB on rice production in field conditions, with some updated information on soil microbial aspects. To study the impact of RHB and CSR-BIO (commercialized bio-formulation), on soil physico-chemical properties, soil microbial biomass (SMB) quantity and paddy productivity, four treatments were set up: control, RHB, CSR-BIO and RHB + CSR-BIO. The RHB with CSR-BIO both the amendments were applied at a rate of 10 t ha−1. Across treatments, the water holding capacity, total -C, -N, -P concentrations and soil moisture content were statistically higher in RHB and CSR-BIO treated soils over the control. The highest SMB-C, -N and -P (408.66 ± 0.57, 83.33 ± 2.08 and 25.66 ± 1.52 μg g−1 dry soil, respectively) was recorded in RHB + CSR-BIO treated soil. Across the sampling dates, SMB-C, -N, -P and inorganic-N (ammonium- and nitrate-N) concentrations were minimum on 35 day after transplantation (DAT) (tillering stage-active growth period), and maximum on 105 DAT (maturity stage). The paddy plant growth variables (panicle length, tiller number, rice grain and paddy straw yields) were found higher in treated plots compared to untreated (control) plots, and varied significantly (P ≤ 0.001) due to treatments. Among the various selected paddy agronomic variables, the application of RHB and CSR-BIO treatment was more pronounced to the yield of rice grains. Results indicate that an increase in the quantity of SMB due to RHB + CSR-BIO addition, improves the soil nutrient status and hence, paddy productivity in nutrient poor agriculture soils. It is suggested that RHB generation from rice husk biochar could be a sustainable crop residues waste management option to enhance the nutrient status, microbial biomass and paddy productivity of disturbed agriculture soils.

Spatial-temporal changes of rainfall erosivity in the loess plateau, China: Changing patterns, causes and implications

Abstract: Rainfall erosivity is one of the key factors influencing soil erosion by water. Improved knowledge of rainfall erosivity is critical for prediction of soil erosion and the implementation of soil and water conservation plan as well as sediment management projects under climate change. In this study, the Jing River Basin (JRB), a typical eco-environmentally vulnerable region of the Loess Plateau in China was selected as a case study. Spatial-temporal changing patterns of rainfall erosivity in the JRB were first examined, followed by detailed investigations of the underlying causes through exploring the relations among annual rainfall, large-scale atmospheric circulation patterns and rainfall erosivity using the cross wavelet technique. Furthermore, implications of changing rainfall erosivity for sediment load and vegetation cover were analyzed. Results indicated that: (1) the year 1985 was a turning point in the time series of annual rainfall erosivity, demonstrating the non-stationary feature. Seasonal rainfall erosivity showed a spatial gradient with decrease from the upper to the lower stream. Rainfall erosivity was the largest in summer, and has increased significantly in the eastern basin; (2) annual rainfall erosivity showed a strong positive correlation with annual rainfall amount, implying that decrease of rainfall may have led to the reduction of rainfall erosivity in recent decades; (3) El Nino-Southern Oscillation and Pacific Decadal Oscillation were correlated with rainfall erosivity during 1982–1991, suggesting that large-scale atmospheric circulation patterns have strong influences on the changing patterns of rainfall erosivity; (4) changing rainfall erosivity had negligible impacts on the variation of vegetation cover (as indexed by the Normalized Differential Vegetation Index), but has detectable influence on sediment discharge which was further modulated by local soil and water conservation practice since the 1970s. These findings are helpful for prediction of soil erosion and adaptation strategies through local soil erosion control measures and sediment control projects.

The role of human activities on sediment connectivity of shallow landslides

Abstract: Sediment connectivity within a catchment depends largely on the morphological complexity of the catchment and is strictly related to the anthropogenic modification of the landscape. In this context, the present research evaluates the role of anthropogenic effects on landscape modifications and the resulting influence on sediment delivery. An assessment of sediment connectivity was carried out for three different human impact scenarios: (i) drainage system density reduction, (ii) road network variation and (iii) land use changes. In addition, shallow landslides were used as sediment source areas to evaluate the potential connection between these sediment sources and downstream areas (e.g. main channels and road network). Two small catchments in the Oltrepo Pavese area (Northern Apennines, Italy), with different size and morphological setting, were analysed: the Rio Frate (1.9 km 2 ) and the Versa (38 km 2 ) catchments. In both areas, several shallow landslides were triggered in 2009 (Rio Frate and Versa) and in 2013 (Versa). Results highlight the role of the landscape complexity in coupling/decoupling upstream sediment sources, such as shallow landslides, from the main channel network and roads. In addition, the analysis identified instability phenomena characterized by high connectivity values, allowing determination of the areas in which mobilized sediment could potentially damage important infrastructures such as the road network or contribute to flooding induced by aggradation or obstruction of the river bed. The proposed approach provides a methodological framework to help improve watershed and land management strategies, especially in shallow landslides prone-areas.

Aggregate stability and associated organic carbon and nitrogen as affected by soil erosion and vegetation rehabilitation on the Loess Plateau

Abstract: Although soil erosion and land use change have long been focuses in carbon research, the combined influence of soil erosion and vegetation rehabilitation on aggregate stability and the associated soil organic carbon (SOC) and total nitrogen (TN) remains unclear. The current study evaluated the effects of soil erosion on aggregate stability and the associated SOC and TN dynamics in relation to vegetation rehabilitation after the implementation of the “Grain-for-Green” project in the hilly Loess region. A check dam sediment sequence was dated using 137Cs activity and erosive rainfall events. The SOC and TN in the bulk soil and aggregate fractions were measured in soils from rehabilitated grasslands and sloping croplands and in sediments retained by the check dam. The results showed that vegetation rehabilitation led to 78%, 27% and 9% average increases in the macroaggregate amount, mean weight diameter (MWD) and mean geometric diameter (MGD), respectively. In addition, rehabilitation resulted in the highest SOC and TN concentrations and contents in macroaggregates among all the aggregate size fractions. Soil erosion facilitated the modification of the aggregate size distributions along with soil mineralization and induced the incorporation of deeper SOC-poor soils during transport. These processes resulted in the aggregate-associated SOC and TN concentrations and contents in the sediments being significantly lower than those in the eroding sloping cropland soils. The highest reductions were found in microaggregates, which exhibited decreases of 48% and 44% for SOC and TN, respectively. Moreover, reaggregation and gully soils incorporated during soil erosion led to higher values of macroaggregate amount and aggregate stability at depositional sites than those at eroding sloping cropland sites in this study. Our study contributes to the understanding of the effects of soil erosion and vegetation rehabilitation on SOC and TN dynamics, which is crucial for understanding the restoration efficiency in soil erosion control and ecosystem security evaluation.

Relationship between soil nutrient properties and biological activities along a restoration chronosequence of Pinus tabulaeformis plantation forests in the Ziwuling Mountains, China

Abstract: The goal of this study was to determine the dynamics of physicochemical properties and biological activities of soil, and their relationship following afforestation. Soil samples were collected at 0–10 cm from five Pinus tabulaeformis plantation forests restored for 15, 25, 30, 45, and 70 years, as well as a millet (Setaria italica) farmland in the Damaiji catchment area. These five afforested lands were converted from similar farmlands. The activities of catalase (CAT), saccharase (SAC), urease (URE) and alkaline phosphatase (ALP), microbial biomass, soil water content, pH, soil bulk density, soil organic carbon (SOC), total nitrogen (TN), ammonium nitrogen, nitrate nitrogen, total phosphorus (TP), and available phosphorus (AP) were measured. The results revealed that the contents of SOC, TN, AP, microbial biomass C, N, and P, CAT, SAC, URE, and ALP in the P. tabulaeformis forest soil were significantly higher than those in the farmland by 64.97%–262.02%, 75.44%–254.28%, 46.63%–114.21%, 125.95%–554.83%, 101.35%–464.21%, 15.80%–167.35%, 22.57%–49.95%, 96.78%–145.73%, 6.98%–56.08% and 89.15%–177.89%, respectively. The soil properties, microbial biomass, enzymatic activities, and C:P and N:P ratios in soil and microbial biomass improved with increasing plantation chronosequence. Variations in C:N:P stoichiometry and higher C:P and N:P ratios in the soil and microbial biomass revealed the P limitation. Simultaneously, N:P ratios included more serviceable information that reflected the relationship between soil and microbes. Soil enzymatic activities had a high correlation with soil nutrient cycling and could be indicators of soil fertility status, particularly for ALP. The significant correlation between SOC, TN, enzymatic activities, and microbial biomass revealed that the substrate availability of carbon and nitrogen could influence the activity of soil enzymes and microorganisms. This study demonstrated that soil enzymatic activities and microorganisms respond to the process of afforestation and hence have the potential to affect nutrient balance and quality of soil.

Positive and negative effects of biochar from coconut husks, orange bagasse and pine wood chips on maize (Zea mays L.) growth and nutrition

Abstract: Transformation of organic waste into biochar for land application is a relatively new green technology management tool. Land-applied biochars can improve soil quality and plant growth. The aim of our study was to investigate the effects of biochars derived from coconut husk, orange bagasse and pine wood chips at different rates of application (0, 5, 10, 20 and 60 t ha− 1), on the biomass, nitrogen (N) and phosphorus (P) status of maize (Zea mays L) cultivated in a sandy soil, under greenhouse conditions. The treatments were arranged in a completely randomized block design with four replications. The effects of biochar addition on plant dry biomass and nutrition were dependent upon the biochar type and application rate. Soil treated with coconut husk biochar at an equivalent rate of 30 t ha− 1 resulted in a 90% increase in maize biomass and plant N and P concentrations of 0.88 and 0.12%, respectively. Orange bagasse biochar applied at a similar rate had no effect on plant biomass, and resulted in plant N and P concentrations of 0.85 and 0.15%, respectively. Application of pine wood chip biochar to soil did not affect plant biomass or nutrition. Even though soil total N increased with an increasing application rate of orange bagasse biochar, N leaching may not have posed a problem since KCl extractable N decreased. However, the associated increase in soil pH may result in potentially greater N losses over time. Thus, the increase in plant biomass and nutrition indicates the superiority of the coconut husk biochar as soil amendment; yet, the application of orange bagasse biochar needs more investigation.

Pedogenic and microbial interrelations to regional climate and local topography: New insights from a climate gradient (arid to humid) along the Coastal Cordillera of Chile

Abstract: The effects of climate and topography on soil physico-chemical and microbial parameters were studied along an extensive latitudinal climate gradient in the Coastal Cordillera of Chile (26°–38°S) The study sites encompass arid (Pan de Azucar), semiarid (Santa Gracia), mediterranean (La Campana) and humid (Nahuelbuta) climates and vegetation, ranging from arid desert, dominated by biological soil crusts (biocrusts), semiarid shrubland and mediterranean sclerophyllous forest, where biocrusts are present but do have a seasonal pattern to temperate-mixed forest, where biocrusts only occur as an early pioneering development stage after disturbance All soils originate from granitic parent materials and show very strong differences in pedogenesis intensity and soil depth Most of the investigated physical, chemical and microbiological soil properties showed distinct trends along the climate gradient Further, abrupt changes between the arid northernmost study site and the other semi-arid to humid sites can be shown, which indicate non-linearity and thresholds along the climate gradient Clay and total organic carbon contents (TOC) as well as Ah horizons and solum depths increased from arid to humid climates, whereas bulk density (BD), pH values and base saturation (BS) decreased These properties demonstrate the accumulation of organic matter, clay formation and element leaching as key-pedogenic processes with increasing humidity However, the soils in the northern arid climate do not follow this overall latitudinal trend, because texture and BD are largely controlled by aeolian input of dust and sea salts spray followed by the formation of secondary evaporate minerals Total soil DNA concentrations and TOC increased from arid to humid sites, while areal coverage by biocrusts exhibited an opposite trend Relative bacterial and archaeal abundances were lower in the arid site, but for the other sites the local variability exceeds the variability along the climate gradient Differences in soil properties between topographic positions were most pronounced at the study sites with the mediterranean and humid climate, whereas microbial abundances were independent on topography across all study sites In general, the regional climate is the strongest controlling factor for pedogenesis and microbial parameters in soils developed from the same parent material Topographic position along individual slopes of limited length augmented this effect only under humid conditions, where water erosion likely relocated particles and elements downward The change from alkaline to neutral soil pH between the arid and the semi-arid site coincided with qualitative differences in soil formation as well as microbial habitats This also reflects non-linear relationships of pedogenic and microbial processes in soils depending on climate with a sharp threshold between arid and semi-arid conditions Therefore, the soils on the transition between arid and semi-arid conditions are especially sensitive and may be well used as indicators of long and medium-term climate changes Concluding, the unique latitudinal precipitation gradient in the Coastal Cordillera of Chile is predestined to investigate the effects of the main soil forming factor – climate – on pedogenic processes

Spatial soil erosion estimation using an artificial neural network (ANN) and field plot data

Abstract: Soil erosion and sediment transport measurement is a time-consuming and difficult step yet important part of hydrological studies. Hence, use of models has become commonplace in estimating soil erosion and sediment transport. In this study, we used an artificial neural network (ANN) to simulate soil erosion rates. A geographic information system (GIS) was used as a pre-processor and post-processor tool to present the spatial variation of the soil erosion rate. The ANN was trained, optimized and verified using data from the Kasilian watershed located in the northern part of Iran. Field plots were used to estimate soil erosion values on the hillslopes. A Multi Layer Perceptron (MLP) network was adopted, where the soil erosion rate was the output variable and the rainfall intensity and amount, air and soil temperature, soil moisture, vegetation cover and slope were the inputs. After the training process, the network was tested. According to the test results, the ANN can estimate soil erosion with an acceptable level (coefficient of determination = 0.94, mean squared error = 0.04). The verified network and its inputs were used to estimate soil erosion rates on the hillslopes. Finally, a soil erosion rate map was generated based on the results of the verified network and GIS capabilities. The results confirm the high potential when coupling an ANN and a GIS in soil erosion estimation and mapping on the hillslopes.

Factors driving the relationships between vegetation and soil properties in the Yellow River Delta, China

Abstract: For wetland ecological restoration, characterizing the relationship between plant communities and soil characteristics has long been recognized as a key issue. Due to the spatial variability in soil properties, multi-scale studies in the context of scale issue is necessary to reveal the complex relationship operating at different intensities. However, insufficient research focuses on the relationship between soil and vegetation at different scales in the Yellow River Delta. In our study, Canonical Correspondence Analysis and the lowest Akaike Information Criteria methods were used to investigate their inherent relationships at three scales (region, sub-region and landscape scale) in Yellow River Delta. The results showed that vegetation properties were strongly related to different variables of soil characteristics at different scales. At the region and landscape scales, soil organic matter, K+ and SO42− were strongly related to vegetation properties, while soil water content, NO3−, soil organic matter and total phosphorus were more important at sub-region scale. Additionally, we found that the soil organic matter was most strongly related to vegetation coverage at the region scale based on the result with the lowest Akaike Information Criterion. Soil nutrients and inorganic ions might be more strongly related to vegetation properties, and their correlations varied according to scale.

Spatial variability of soil texture fractions and pH in a flood plain (case study from eastern Iran)

Abstract: Flood plain ecosystems show significant soil spatial variability. Understanding spatial variations of soil texture fractions and pH in flood plains is necessary for ensuring proper management of these plains, because these properties influence soil structure, fertility, hydraulic conductivity, infiltration, and erosion. In the present study, the distribution of soil texture fractions and pH was investigated in a flood plain with intensive wind erosion for an area of ~ 41,000 ha in Zahak county of Sistan and Baluchestan province in eastern Iran. A random forest technique was used to link environmental variables and the studied properties. 460 soil samples were collected from 0 to 30 cm depth across a 750 m grid. 361 samples were used for training and 99 for independent validation. Results showed that the distance from the river was the most important environmental variable for predicting soil texture fractions and pH in the study area. Natural channel networks, elevation, valley depth, LS factor, NDSI, vertical distance to channel networks, slope, wind effect, NDVI, and brightness were other important variables. The maps produced indicated a higher sand content near Sistan River. Clay, silt, and pH contents increased with distance from Sistan River. Results showed that clay and pH had a similar distribution in the study area. The values of RMSE for the maps of estimated sand, silt, clay, and pH in validation data were respectively 21.40, 17.45, 6.06 and 0.45. These values of RMSE for sand, silt, clay, and pH were respectively 10.3, 10.7, 15.1, and 13.3% lower than a simple model (mean model). Results indicated that using the distance from the river and channel networks as a variable in digital soil mapping can increase the accuracy of the predictive maps of soil properties in flood plains.

A comparative study of an expert knowledge-based model and two data-driven models for landslide susceptibility mapping

Abstract: In this study, an expert knowledge-based model, a logistic regression model, and an artificial neural network model were compared for their accuracy and portability in landslide susceptibility mapping. Two study areas (the Kaixian and the Three Gorges areas in China) were selected for this comparison based on their well-known, high landslide hazard. To evaluate the performance of these models and to minimize the impact of the particularity of a study area on model performance, cross-applications of three models between the two study areas were conducted. When the Kaixian area was used as a model development area, prediction accuracy for the expert knowledge-based model, the logistic regression model, and the artificial neural network model were 71.5%, 81.0% and 100.0%, respectively. The high prediction accuracy of the two data-driven models were expected because the observed landslide occurrence used in training the models were also used to validate their respective performance, while the expert knowledge-based model did not use these observations for training. The perfect accuracy for the neural network model can also be attributed to its over-prediction of the susceptibility. When breaking the susceptibility into four classes: very low susceptibility (0–0.25), low susceptibility (0.25–0.5), high susceptibility (0.5–0.75), and very high susceptibility (0.75–1), the observed landslide density at the very high susceptibility level is 0.303/km2, 0.212/km2, and 0.195/km2 for the expert knowledge-based model, the logistic regression model, and the artificial neural network model, respectively. This suggests that the expert knowledge-based model was much better than the other two data-driven models at evaluating landslide occurrence in very high susceptibility areas. When the three models developed in the Kaixian area were applied in the Three Gorges area without any changes, their prediction accuracy dropped to 44.8% for the logistic regression model and 81.6% for the artificial neural network model, while the expert knowledge-based model maintained its initial accuracy level of 82.8%. The landslide density for the very high susceptibility areas in the Three Georges area was 0.275/km2, 0.082/km2, and 0.060/km2 for the expert knowledge-based model, the logistic model, and the artificial neural network model, respectively. These results indicate that the expert knowledge-based model is more effective at predicting areas with very high susceptibility. When the Three Gorges area was used as a model development area and the Kaixian area was used as the model application area, similar results were obtained. Results from the two experiments show that the performance of the logistic regression model and artificial neural network model is not as stable as the expert knowledge-based model when transferred to a new area. This suggests that the expert knowledge-based model is more suitable for landslide susceptibility mapping over large areas.

Soil aggregate stability under different rain conditions for three vegetation types on the Loess Plateau (China)

Abstract: Soil aggregate stability is essential for moderating the soil quality and preventing soil erosion. Vegetation restoration may effectively increase the stability of soil aggregates via soil organic matter. This study was designed to investigate the effects of vegetation types with long-term revegetation on the soil aggregate characteristics. Three vegetation type zones (grass land, forest-grass land and forest land) were selected in the Yanhe Watershed (northwest China) as the subjects. Soil aggregate stability was determined by the method of Le Bissonnais, including three disruptive tests: fast wetting (FW), slow wetting (SW) and mechanical breakdown (WS). The results showed that the mean weighted diameter (MWD) significantly differed from the tests and vegetation types. In the 0–10 cm soil layer, MWD ranged from 2.65 to 3.26 mm for the SW test, which corresponded to very stable soil aggregate; they ranged from 0.53 to 1.08 mm for the WS test, and from 0.57 to 1.96 mm for the FW test, both of which corresponded to very unstable soil aggregates. In the 10–20 cm soil layer, MWD ranged from 2.75 to 3.33 mm for the SW test, 0.39 to 0.83 mm for the WS test, and 0.44 to 1.37 mm for the FW test. The MWDs under the three tests were the lowest for the grass land at both soil layers, and the MWDs for the WS and FW tests were significantly lower than the MWD for the SW test. In all three tests, MWDs showed the same order: forest land > forest-grass land > grass land. MWD indicated that forest land had much stronger ability to resist soil erosion no matter the rain conditions. The correlations between soil organic matter content and MWD for the FW and WS tests were significant (P

Comparison of the presence-only method and presence-absence method in landslide susceptibility mapping

Abstract: Presence-absence methods are widely-used data-driven models for landslide susceptibility mapping Landslide absence data included in the training data of presence-absence methods is usually not available and has to be generated In consideration of low availability and uncertain quality of landslide absence data, many presence-only methods which simply use landslide presence as training data were proposed to map landslide susceptibility However, whether the presence-only methods can circumvent the influence of the shortcomings inherent to landslide absence data and perform better than presence-absence methods are worth studying Moreover, the effect of landslide absence data in data-driven models for landslide susceptibility mapping can be discussed In this study, two presence-only methods including one-class support vector machine (one-class SVM), kernel density estimation (KDE), and two presence-absence methods including artificial neural networks (ANN) and two-class support vector machine (two-class SVM) are developed and compared to evaluate their respective performance in mapping landslide susceptibility The AUC values are 0705, 0720, 0929, and 0951 for one-class SVM, KDE, ANN, and two-class SVM, respectively From the comparison of the four methods, two-class SVM has the best performance in landslide susceptibility mapping among the four methods, while one-class SVM has the worst Two presence-absence methods can constrain the over-prediction of susceptibility value better and have better performance than the two presence-only methods since they classify less percentage of areas to be susceptible with more landslide occurrences located inside The landslide absence data is proven to constrain the over-prediction of models, which makes it necessary in landslide susceptibility mapping

Chemistry and microbiology of the Critical Zone along a steep climate and vegetation gradient in the Chilean Coastal Cordillera.

Abstract: The Chilean Coastal Cordillera features a spectacular climate and vegetation gradient, ranging from arid and unvegetated areas in the north to humid and forested areas in the south. The EarthShape project (“Earth Surface Shaping by Biota”) uses this natural gradient to investigate how climate and biological processes shape the Earth's surface. We explored the Critical Zone, the Earth's uppermost layer, in four key sites located in desert, semidesert, Mediterranean, and temperate climate zones of the Coastal Cordillera, with the focus on weathering of granitic rock. Here, we present first results from 16 approximately 2 m-deep regolith profiles to document: (1) architecture of weathering zone; (2) degree and rate of rock weathering, thus the release of mineral-derived nutrients to the terrestrial ecosystems; (3) denudation rates; and (4) microbial abundances of bacteria and archaea in the saprolite. From north to south, denudation rates from cosmogenic nuclides are ~10 t km−2 yr−1 at the arid Pan de Azucar site, ~20 t km−2 yr−1 at the semi-arid site of Santa Gracia, ~60 t km−2 yr−1 at the Mediterranean climate site of La Campana, and ~30 t km−2 yr−1 at the humid site of Nahuelbuta. A and B horizons increase in thickness and elemental depletion or enrichment increases from north (~26°S) to south (~38°S) in these horizons. Differences in the degree of chemical weathering, quantified by the chemical depletion fraction (CDF), are significant only between the arid and sparsely vegetated site and the other three sites. Differences in the CDF between the sites, and elemental depletion within the sites are sometimes smaller than the variations induced by the bedrock heterogeneity. Microbial abundances (bacteria and archaea) in saprolite substantially increase from the arid to the semi-arid sites. With this study, we provide a comprehensive dataset characterizing the Critical Zone geochemistry in the Chilean Coastal Cordillera. This dataset confirms climatic controls on weathering and denudation rates and provides prerequisites to quantify the role of biota in future studies.