Soil Type Classification Research Articles

A dataset of Gansu soil species based on the Second National Soil Survey

Based on the results of the Second National soil Survey, Gansu Soil Species was established to accumulate systematic, standardized and quantitative data of soil classification as a solid foundation for the development of soil science in Gansu province.This dataset is based on the processing and integration of Gansu Soil Species published by Gansu Science and Technology Press in 1992. The original data and information of 285 soil species in the book (e.g. attribution and distribution, main characters, production performance summary, physical and chemical properties of profiles) are extracted and collected in a standardized, structured and digital way, integrated with 8 specialized subject data tables indexed by species ID and species name, and conducive to the development of online query, including soil species classification, distribution and characteristics of main soil types, basic information and distribution of soil species, sampling sites and occurrence layers of typical profiles, statistical physicochemical properties of typical profiles and soil species profiles, etc. This dataset can be widely used in regional land quality assessment, environmental effect research, soil carbon storage research, etc., providing significant scientific support for the promotion of new agricultural technology and rational utilization of soil resources.

Hyperspectral Image Classification of Soils using Soft Computing Techniques

In geotechnical engineering classification of soils is an important accept because to provide guidance about soils based on their experimental analysis. The classification of system can be done in various ways because one should not rely on empirical information for solving the problems. These soil types are classified based on their behavioural conditions and based on generalized information provided by the scientists, engineers and resource persons. Classification of soil type by visiting particular location is difficult so remote sensing images are used. The most widely used remote sensing method for classification of soils is hyperspectral image (HSI) technology. To classify HSI different soft computing techniques are applied and compared algorithms with their accuracies to know the best algorithm for classification of soil types.

Historical Global Review of Acid-Volatile Sulfide Sediment Monitoring Data

Acid-volatile sulfides (AVS) are strongly associated with the bioavailability of some divalent metals such as cadmium, copper, lead, nickel and zinc. However, the global spatial variability of AVS for aquatic systems is unknown. The specific goals of this study were to: (1) summarize all available AVS monitoring data from all types of freshwater and saltwater waterbodies (streams/creeks, rivers, lakes/ponds/reservoirs and estuarine/marine areas) and (2) compare AVS concentrations from these various types of waterbodies considering both soil type classification and biomes. AVS measurements were reported from 21 different countries. A total of 17 different soil types were reported for all waterbody types and both podzols and luvisols were found in all waterbody types. Nine different biomes were sampled for all waterbody types. The temperate broadleaf and mixed forest biome was sampled for AVS in all waterbody types. Mean AVS concentrations ranged from 0.01 to 503 µmoles/g for 140 different waterbody types and the 90th centile for all these waterbodies was 49.4 µmoles/g. A ranking of waterbody type means from low to high AVS measurements showed the lowest mean value was reported for streams/creeks (5.12 µmoles/g; range from 0.1 to 39.8 µmoles/g) followed by lakes/ponds/reservoirs (11.3 µmoles/g; range from 0.79 to 127 µmoles/g); estuarine/marine areas (27.2 µmoles/g; range from 0.06 to 503 µmoles/g) and rivers (27.7 µmoles/g; range from 1.13 to 197 µmoles/g). The data provided in this study are compelling as it showed that the high variability of AVS measurements within each waterbody type as well as the variability of AVS within specific locations were often multiple orders of magnitude differences for concentration ranges. Therefore, a comprehensive spatial and temporal scale sampling of AVS in concert with divalent metals analysis is critical to avoid possible errors when evaluating the potential ecological risk of divalent metals in sediment.

Cone penetration index for soil behaviour type prediction

Several approaches have been proposed to classify soil types using both cone and piezocone penetration tests. The chart relating the normalized cone resistance (Qtn) and the friction ratio (Fr) has proven to be the most reliable method. However, its practical use requires function fitting, where the chart is described by means of a soil behaviour type index. The currently available indexes for this chart, based on concentric circles and hyperbolas, lead to significant errors. Thus, to properly represent the Qtn–Fr chart, this study proposes new soil behaviour type indexes based on an analytical approach from a concentric logarithmic spiral approximation and a numerical approach from an exponential function approximation. The new indexes provide an improvement in the reproduction of the soil type zones while preserving the robustness of the original method. A discussion on the cone penetration test-based soil type classification is presented, along with a case example comparing the soil behaviour indexes in soil profiling.

Relation of shear wave velocity variations with depth for different lithologies: A contribution towards mitigating the region's seismic risk

The state of Uttarakhand in India lies in a highly seismically active zone, therefore earthquake re- sistant design criteria have been followed for major construction practice in this region. One of the major inputs for earthquake-resistant design is average shear wave velocity (VS ) at 30 m depth. In the present work, at twenty-six different sites, the Horizontal to vertical spectral ratio (HVSR) technique and multichannel analysis of surface wave (MASW) is used for joint analysis to obtain a one-dimensional velocity model that supports both the HV spectrum and dispersion curve obtained from HVSR and MASW methods, respectively. Shear wave profiles obtained at various sites are compared with the bore log data obtained from drilling and it clearly shows that the (VS ) for the same formation have a strong dependence on the depth at which it occurred. Data from twenty- four shear wave profiles have been used to prepare a linear regression relation of (VS ) for different lithological formations with respect to their depth of occurrence. Root mean square error obtained from the developed relationship for various lithologies clearly shows that the results are fairly well within the range of acceptance. The developed relations have been further validated by calculating Vs profile from bore log data obtained at two new location sites that are not included in the data set used for the preparation of regression relations. A comparison of two velocity sections clearly shows that the velocity profile computed from regression relation matches closely with that obtained from the seismic survey and thereby establishing the efficacy of developed regression relationship for its practical implementation. This developed relation allows soil type classification for zoning purposes.

Mapping of Land Cover with Optical Images, Supervised Algorithms, and Google Earth Engine.

Crops and ecosystems constantly change, and risks are derived from heavy rains, hurricanes, droughts, human activities, climate change, etc. This has caused additional damages with economic and social impacts. Natural phenomena have caused the loss of crop areas, which endangers food security, destruction of the habitat of species of flora and fauna, and flooding of populations, among others. To help in the solution, it is necessary to develop strategies that maximize agricultural production as well as reduce land wear, environmental impact, and contamination of water resources. The generation of crop and land-use maps is advantageous for identifying suitable crop areas and collecting precise information about the produce. In this work, a strategy is proposed to identify and map sorghum and corn crops as well as land use and land cover. Our approach uses Sentinel-2 satellite images, spectral indices for the phenological detection of vegetation and water bodies, and automatic learning methods: support vector machine, random forest, and classification and regression trees. The study area is a tropical agricultural area with water bodies located in southeastern Mexico. The study was carried out from 2017 to 2019, and considering the climate and growing seasons of the site, two seasons were created for each year. Land use was identified as: water bodies, land in recovery, urban areas, sandy areas, and tropical rainforest. The results in overall accuracy were: 0.99% for the support vector machine, 0.95% for the random forest, and 0.92% for classification and regression trees. The kappa index was: 0.99% for the support vector machine, 0.97% for the random forest, and 0.94% for classification and regression trees. The support vector machine obtained the lowest percentage of false positives and margin of error. It also acquired better results in the classification of soil types and identification of crops.

A novel PCA-based calibration algorithm for classification of challenging laser-induced breakdown spectroscopy soil sample data

Accurate classification of soil types and contamination is crucial for crops' productivity. Among the soil analysis techniques, laser-induced breakdown spectroscopy (LIBS) has become a prominent technology for real-time characterization of soil properties. LIBS coupled with supervised machine learning and chemometrics methods (e.g., partial least squares discriminate analysis (PLS-DA), principal component analysis (PCA)) has demonstrated great capabilities for soils classification. However, when the training and test spectra have different distribution and not representative of each other, there are generalization issues, which make the model trained on training spectra hard to adapt to test spectra. In this work we propose a method to calibrate the test spectra using the median of principal components (PCs). PCA is used to analyze the spectra distribution. We independently compute the median of both training's and test's PCs, and then the test's median is adjusted based on its differences with training's. With the calibrated PCs, the test spectra is reconstructed accordingly. To test the performance of the proposed calibration algorithm, we conduct experiments on a publicly available challenging LIBS dataset. We compare our calibration algorithm with the current best performing calibration method on the same test set, using the same machine learning (ML) algorithm, PLS-DA, trained with the same training set. Our method improves the test accuracy by 1.2%. The reason using PLS-DA for performance comparison is that it is currently the best performing ML algorithm. To further improve the test accuracy, other ML algorithms are investigated. Convolutional neural networks (CNN) have achieved good accuracy in lithological classification with LIBS recently. Therefore, it is extended in this work to soil classification. We use CNN as a tool for feature extraction and as an end-to-end classifier. We use the CNN based extraction mechanism with other classifiers, such as support vector machine (SVM) and random forest (RF), for soil classification. The performance of CNN models on the calibrated test spectra is compared, which concludes that CNN combined with SVM achieves the best accuracy and improves the test accuracy by 3.1% compared to the best performing ML algorithm PLS-DA.

Seismic fragility of legged liquid storage tanks based on soil type classifications

Recent earthquakes around the world have caused extensive damage or the complete collapse of legged storage tanks. The damage typically observed includes local buckling in the upper zone of the tank legs and anchor failures and indicates that legged storage tanks are highly vulnerable structures. Although the seismic response of legged storage tanks is affected by structural and hydrodynamic properties, the soil type where the tank is constructed can significantly influence the seismic response of these structures. However, the influence of soil type on the seismic fragility of legged tank structures has been not been extensively studied. In this investigation, the influence of soil type classification on the seismic fragility of legged wine storage tanks is evaluated by comparing the fragility curves of a typical legged storage tank, considering different soil types and different plate thicknesses for the tank legs. A mathematical model capable of capturing the most common failure modes observed was developed and used to generate fragility curves from results of incremental dynamic analyses with the ground motion records grouped by soil type. The fragility curves showed that the probability of failure of the legged storage tanks were highly dependent on the soil type on which the tank was located. Recommended thicknesses of the tank legs according to the soil type, seismic zone and the probability of failure are presented.

A novel feature based algorithm for soil type classification

Agriculture is the backbone of Bangladesh’s economy and it is one of the largest employment sectors. In Bangladesh, the population is increasing rapidly and at the same time, the total cultivable land is decreasing significantly. To ensure maximum crop production using the limited land resources, it is essential to identify and select the appropriate type of soil because different crops need different soil types. Currently, there are two types of methods available to determine the soil type, namely chemical and image analysis. Although the first one is accurate, it is expensive and time consuming. On the other hand, image based soil classification is much cheaper and faster but its accuracy level is low. In this study, we present a novel feature based algorithm that combines quartile histogram oriented gradients (Q-HOG), most frequent varphi -Pixels and a new feature selection method for classifying soil types. We have used four machine learning algorithms and evaluated the performance with different sets of features. We have also compared our work with two prominent and recent works on image-based soil classification systems. The experimental results show that the performance of our proposed method in terms of four standard evaluation metrics, namely accuracy, precision, F1_score, and recall scores are higher than the existing image-based soil classification systems.

Evaluation of soil replacement and cement grout injection in soil settlement

This paper discusses the process of original soil stabilization in Trans Sumatra Bakauheni-Terbanggi Besar Toll Road Project Package 2 Sidomulyo-Kotabaru. The soil replacement process was conducted at approximately 24 kilometres along the toll’s main road. The original soil bearing capacity analysis stage was by performing a Dynamic Cone Penetrometer (DCP) and Sondir test to analyze the deep of hard soil. A soil replacement was carried out to replace the original soil with soil that has appropriate specification. the piling up process was conducted in stages, which has Sandstone in such of the layer. The research done about the landfill sample was retaken and collected at 68 points. The stockpile soil samples collection was then followed by the analysis which was conducted in the laboratory to find the soil bearing capacity. There are 4 types of bearing capacity parameters analyzed, namely specific gravity, water content, aggregate analysis (Sieve Analysis), and consistency limit (Atterberg Limit). Referring to it, then there was the classification of soil types according to AASHTO M145 & Casagrande Soil Classification System. A point that has settlement after soil replacement is STA 52+000. So, there need to add soil stabilization, that is cement grout injection. Researchers analyzed the soil settlement by interpretation method. Results showed that soil replacement and cement grout injection could decrease a soil settlement by about 15.07 cm to become 0.93 cm.

Effective Soil Type Classification Using Convolutional Neural Network

Soil classification is a growing research area in the current era. Various studies have proposed different techniques to deal with the issues, including rule-based, statistical, and traditional learning methods. However, the plans remain drawbacks to producing an accurate classification result. Therefore, we propose a novel technique to address soil classification by implementing a deep learning algorithm to construct an effective model. Based on the experiment result, the proposed model can obtain classification results with an accuracy rate of 97% and a loss of 0.1606. Furthermore, we also received an F1-score of 98%.

Geomorphology-Wavelet based approach to rainfall runoff modeling for data scarce semi-arid regions, Kolar River catchment, India

The conceptual and physical mathematical model of rainfall-runoff modeling uses various parameters such as land use land cover, soil type classification, rainfall, atmospheric data such as temperature, evapotranspiration, solar radiation and wind speed, etc. But these data may not be available for developing countries and data scares semi-arid watershed. Also, the problem is even more critical for ungauged catchments and where manual record is maintained of water level and rainfall data. To address this issue, trend analysis is performed using Mann-Kendall test and Sen’s slope test which shows significant trend change stressing the need for new method for runoff prediction for better water resource management. In this study, a total of four models namely nonlinear autoregressive model with exogenous inputs lumped (LNARX), nonlinear autoregressive model with exogenous geomorphometrically processed inputs (GNARX), wavelet nonlinear autoregressive model with exogenous inputs (WLNARX) and nonlinear autoregressive model with exogenous geomorphometrically processed inputs (WGNARX). Ten models with different input combinations were selected based on their performance are analyzed for all the four networks. The best performing model for these networks is model no. 6 with WGNARX network with NSE 0.97 and RMSE 0.97 and with least value of RMSE. This method can be applied to data scarce region where data available are available for shorter duration and helpful for ungauged catchments also.

Drivers of Organic Carbon Stocks in Different LULC History and along Soil Depth for a 30 Years Image Time Series

Soil organic carbon (SOC) stocks are a remarkable property for soil and environmental monitoring. The understanding of their dynamics in crop soils must go forward. The objective of this study was to determine the impact of temporal environmental controlling factors obtained by satellite images over the SOC stocks along soil depth, using machine learning algorithms. The work was carried out in São Paulo state (Brazil) in an area of 2577 km2. We obtained a dataset of boreholes with soil analyses from topsoil to subsoil (0–100 cm). Additionally, remote sensing covariates (30 years of land use history, vegetation indexes), soil properties (i.e., clay, sand, mineralogy), soil types (classification), geology, climate and relief information were used. All covariates were confronted with SOC stocks contents, to identify their impact. Afterwards, the abilities of the predictive models were tested by splitting soil samples into two random groups (70 for training and 30% for model testing). We observed that the mean values of SOC stocks decreased by increasing the depth in all land use and land cover (LULC) historical classes. The results indicated that the random forest with recursive features elimination (RFE) was an accurate technique for predicting SOC stocks and finding controlling factors. We also found that the soil properties (especially clay and CEC), terrain attributes, geology, bioclimatic parameters and land use history were the most critical factors in controlling the SOC stocks in all LULC history and soil depths. We concluded that random forest coupled with RFE could be a functional approach to detect, map and monitor SOC stocks using environmental and remote sensing data.

Measurement of Multichannel Analysis of Surface Wave in the South of Pangandaran as an Effort for Earthquake Hazard Mitigation

The Regency of Pangandaran is located on the southern side of West Java, surrounded by local faults, and close to the Indo-Australia and Eurasian subduction zone, causing the area to feel the earthquake’s impact often. Tasikmalaya Earthquake M 7.3, on September 2, 2009, was felt strong and has caused damage in this area. Therefore, mitigation efforts are critical to reducing the earthquake impact. This study aims to determine the classification of soil types in the southern Pangandaran based on 1D Vs up to 30 m depth (Vs30) measured by the Multichannel Analysis of Surface Wave (MASW) method. The study used the SNI 1726-2012 as the categorization of soil types based on Vs30. To clarify, the Vs profile made the cross-section from west to east part in 2D. The results show that hard and medium soils dominate the research area with Vs30 values between 210 - 627 m/s that make the area relatively safe from the potential of strong earthquake shocks. However, the southern part of Cijulang, Parigi, and Pangandaran districts have medium soil types that are relatively thick. Those parts may indicate the potential for stronger earthquake shocks.

Digital soil maps can perform as well as large-scale conventional soil maps for the prediction of catchment baseflows

The rapid advance of digital soil mapping (DSM) has resulted in the generation of fine resolution soils spatial datasets with associated uncertainty that can be as accurate but more cost-effective and faster to produce than conventional soil maps. There is documented and increasing interest by policy makers and end-users in moving from conventional soil mapping to DSM; however, the wider operational use of DSM depends on demonstrating the effectiveness of DSM in fulfilling user needs and requirements along with providing DSM products that can be easily used by non-specialists. In this study we used the Hydrology of Soil Types (HOST) classification scheme, which was devised to predict flows in ungauged catchments and is used by both the research and policy maker communities in the UK, as the exemplar for comparing the efficiency of HOST class maps produced using both DSM and conventional soil map approaches for predicting catchment hydrological response. The performance for hydrological predictions of a detailed (1:25,000 scale) polygon HOST class map and a HOST-DSM class map produced via spatial disaggregation was assessed by comparing Base Flow Index (BFI) calculated using HOST class proportions from both maps with BFI calculated using flow data from gauges in 39 selected catchments. Results showed that the disaggregated HOST-DSM class map gave similar or even better BFI predictions than the conventional polygon-based HOST class map, while also providing a better insight on spatial soil variability within map units and its effect on hydrological predictions. This study demonstrates the potential of DSM to produce soil hydrological maps that provide comparable or better baseflow predictions than maps produced using detailed and intensive soil surveying. The study results also suggest that translating DSM of classes to more easily interpretable mapping datasets and incorporating prediction uncertainty in the final DSM product, as in the case of BFI maps based on HOST-DSM classes, could help with facilitating the transition towards a wider operational use of DSM.

Data Mining Classification Algorithms for Analyzing Soil Data

Rapid changes are occurring in our global ecosystem, and stresses on human well-being, such as climate regulation and food production, are increasing, soil is a critical component of agriculture. The project aims to use Data Mining (DM) classification techniques to predict soil data. Analysis DM classification strategies such as k-Nearest-Neighbors (k-NN), Random-Forest (RF), Decision-Tree (DT) and Naïve-Bayes (NB) are used to predict soil type. These classifier algorithms are used to extract information from soil data. The main purpose of using these classifiers is to find the optimal machine learning classifier in the soil classification. in this paper we are applying some algorithms of DM and machine learning on the data set that we collected by using Weka program, then we compare the experimental result with other papers that worked like our work. According to the experimental results, the highest accuracy is k-NN has of 84 % when compared to the NB (69.23%), DT and RF (53.85 %). As a result, it outperforms the other classifiers. The findings imply that k-NN could be useful for accurate soil type classification in the agricultural domain.

Citizen science with colour blindness: A case study on the Forel-Ule scale.

Many citizen science projects depend on colour vision. Examples include classification of soil or water types and biological monitoring. However, up to 1 in 11 participants are colour blind. We simulate the impact of various forms of colour blindness on measurements with the Forel-Ule scale, which is used to measure water colour by eye with a 21-colour scale. Colour blindness decreases the median discriminability between Forel-Ule colours by up to 33% and makes several colour pairs essentially indistinguishable. This reduces the precision and accuracy of citizen science data and the motivation of participants. These issues can be addressed by including uncertainty estimates in data entry forms and discussing colour blindness in training materials. These conclusions and recommendations apply to colour-based citizen science in general, including other classification and monitoring activities. Being inclusive of the colour blind increases both the social and scientific impact of citizen science.

Implication of classification of land types on the river banks on the layout of settlements planning (A case of the Ogan Ulu community, in South Sumatera, Indonesia)

ABSTRACT This article analyses the implications of the knowledge of the Ogan community in South Sumatra (Indonesia) about the types of land on the banks of the river on the layout of their settlements. Living on the river bank is an important identity for the Ogan Ulu community, but the condition of their rivers which tend to run watery and the inappropriate land use, of course, will have an impact on the destruction of settlements and their natural environment. Through anthropological studies, the results of the study show that the use of land on the river banks are not arbitrary, but it is based on how the community interprets the classification of soil types and the river’s behaviour in the river environment. This article confirms that the river's behaviour and classification of soil types have implications for the community's struggle to maintain settlements as their identity while preserving the river's environment.

A Study of Natural Radioactivity Levels and Radon/Thoron Release Potential of Bedrock and Soil in Southeastern Ireland.

Radon (222Rn) and thoron (220Rn) account for almost two-thirds of the annual average radiation dose received by the Irish population. A detailed study of natural radioactivity levels and radon and thoron exhalation rates was carried out in a legislatively designated “high radon” area, as based on existing indoor radon measurements. Indoor radon concentrations, airborne radiometric data and stream sediment geochemistry were collated, and a set of soil samples were taken from the study area. The exhalation rates of radon (E222Rn) and thoron (E220Rn) for collected samples were determined in the laboratory. The resultant data were classified based on geological and soil type parameters. Geological boundaries were found to be robust classifiers for radon exhalation rates and radon-related variables, whilst soil type classification better differentiates thoron exhalation rates and correlated variables. Linear models were developed to predict the radon and thoron exhalation rates of the study area. Distribution maps of radon and thoron exhalation rates (range: E222Rn [0.15–1.84] and E220Rn [475–3029] Bq m−2 h−1) and annual effective dose (with a mean value of 0.84 mSv y−1) are presented. For some parts of the study area, the calculated annual effective dose exceeds the recommended level of 1 mSv y−1, illustrating a significant radiation risk. Airborne radiometric data were found to be a powerful and fast tool for the prediction of geogenic radon and thoron risk. This robust method can be used for other areas where airborne radiometric data are available.

Ensemble classifier to support decisions on soil classification

Soil performs a significant role in the agricultural ecosystem by supplying essential nutrients and a conducive environment for plants’ growth and crop yield. Inside the agribusiness space, the soil classification is a crucial work that gives good classification results for different soil types. The taxonomy provides an excellent rating for inherent soil elements. This work investigates the accuracy of three well-known classification models like K-Nearest Neighbor (k-NN), Naive Bayes (NB) and, Decision Tree (DT) using a publically available agricultural soil dataset. Post investigation, an Ensemble Classifier (EC) is proposed by fusing the above mentioned three classifiers. The experimental results indicate that EC has the highest accuracy of 84% in comparison to the NB (72.90%), k-NN (73.56%), and DT (80.84%). So it performs better than the other classifiers. The results infer that EC would be useful for accurate classification of soil types in the agricultural domain.