Digital mapping of soil properties using multiple machine learning in a semi-arid region, central Iran

Zeraatpisheh, Mojtaba; Ayoubi, Shamsollah; Jafari, Azam; Tajik, Samaneh; Finke, Peter

doi:10.1016/j.geoderma.2018.09.006

Cited by 230 publications

(97 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For more information about tree learners in soil science for regression, Hengl et al (2017) found lower R 2 using XGB than RF on a global-scale prediction. Zeraatpisheh et al (2018) put forward the lowest RMSE and the highest R 2 using RF compared with multiple linear regression and regression trees for the prediction of clay, and this conclusion was similar to our study.…”

Section: The Systematic Comparison Of the Five Machine Learning Modelssupporting

confidence: 88%

Systematic comparison of five machine-learning methods in classification and interpolation of soil particle size fractions using different transformed data

Zhang¹,

Shi²

2019

Preprint

View full text Add to dashboard Cite

Abstract. Soil texture and soil particle size fractions (psf) play an increasing role in physical, chemical and hydrological processes. Digital soil mapping using machine-learning methods was widely applied to generate more detailed prediction of qualitative or quantitative outputs than traditional soil-mapping methods in soil science. As compositional data, interpolation of soil psf combined with log ratio approaches was developed to improve the prediction accuracy, which also can be used to indirectly derive soil texture. However, few reports systematically analyzed and compared the classification and regression, the accuracies of original (untransformed) and log ratio approaches, and the performance of direct and indirect soil texture classification using machine-learning methods. In this total, a total of 45 evaluation models generated from five different machine-learning models combined with original and three log ratio approaches–additive log ratio, centered log ratio and isometric log ratio (ALR, CLR and ILR, respectively), to evaluate and compare the performance of soil texture classification and soil psf interpolation. The results demonstrated that log ratio approaches modified the soil sampling data more symmetrically, and with respect to soil texture classification, random forest (RF) and extreme gradient boosting (XGB) showed notable consequences. For soil psf interpolation, RF delivered the best performance among five machine-learning models with lowest root mean squared error (RMSE, sand: 15.09 %, silt: 13.86 %, clay: 6.31 %), mean absolute error (MAE, sand: 10.65 %, silt: 9.99 %, clay: 5.00 %), Aitchison distance (AD, 0.84) and standardized residual sum of squares (STRESS, 0.61), and highest coefficient of determination (R2, sand: 53.28 %, silt: 45.77 %, clay: 53.75 %). STRESS was improved using log ratio approaches, especially CLR and ILR. There is a pronounced improvement (21.3 %) in the kappa coefficient using indirect soil texture classification compared to the direct approach. Our systematic comparison helps to elucidate the processing and selection of compositional data in spatial simulation.

show abstract

Section: The Systematic Comparison Of the Five Machine Learning Modelssupporting

confidence: 88%

Systematic comparison of five machine-learning methods in classification and interpolation of soil particle size fractions using different transformed data

Zhang¹,

Shi²

2019

Preprint

View full text Add to dashboard Cite

show abstract

“…The NDVI values range from −0.5 to more than 0.8, indicating a high diversity of vegetation cover spread across the province that has a potentially significant effect on SOC content due to large differences in the number of falling leaves and plant residues. The MrVBF shows that flat valley bottoms where sediments and outflows accumulate leading to higher clay and SOC contents [34,67]. All environmental variables which did not conform with SOC grid resolution of 30 × 30 m were resampled to a 30 m spatial resolution using either the nearest neighbor or bilinear resampling methods.…”

Section: Auxiliary Variablesmentioning

confidence: 99%

Predicting and Mapping of Soil Organic Carbon Using Machine Learning Algorithms in Northern Iran

et al. 2020

View full text Add to dashboard Cite

Estimation of the soil organic carbon (SOC) content is of utmost importance in understanding the chemical, physical, and biological functions of the soil. This study proposes machine learning algorithms of support vector machines (SVM), artificial neural networks (ANN), regression tree, random forest (RF), extreme gradient boosting (XGBoost), and conventional deep neural network (DNN) for advancing prediction models of SOC. Models are trained with 1879 composite surface soil samples, and 105 auxiliary data as predictors. The genetic algorithm is used as a feature selection approach to identify effective variables. The results indicate that precipitation is the most important predictor driving 14.9% of SOC spatial variability followed by the normalized difference vegetation index (12.5%), day temperature index of moderate resolution imaging spectroradiometer (10.6%), multiresolution valley bottom flatness (8.7%) and land use (8.2%), respectively. Based on 10-fold cross-validation, the DNN model reported as a superior algorithm with the lowest prediction error and uncertainty. In terms of accuracy, DNN yielded a mean absolute error of 0.59%, a root mean squared error of 0.75%, a coefficient of determination of 0.65, and Lin’s concordance correlation coefficient of 0.83. The SOC content was the highest in udic soil moisture regime class with mean values of 3.71%, followed by the aquic (2.45%) and xeric (2.10%) classes, respectively. Soils in dense forestlands had the highest SOC contents, whereas soils of younger geological age and alluvial fans had lower SOC. The proposed DNN (hidden layers = 7, and size = 50) is a promising algorithm for handling large numbers of auxiliary data at a province-scale, and due to its flexible structure and the ability to extract more information from the auxiliary data surrounding the sampled observations, it had high accuracy for the prediction of the SOC base-line map and minimal uncertainty.

show abstract

“…O interesse em certificar os resultados compreende o processo de reconhecer a ocorrência de diferentes tipos de solos a partir de três bases fundamentais: posição do relevo (Christofoletti, 1980;Hall, 1983;Miller e Shaetzl, 2015;Shaetzl e Miller, 2016;Marques et al, 2018;Zeraatpisheh et al, 2018), rede de drenagem (Way, 1973;França e Demattê, 1990;…”

Section: Sigla Descrição Referênciaunclassified

Compartimentação da paisagem via relevo e rede de drenagem e sua relação com atributos e classes de solos

Mello¹,

Demattê²

View full text Add to dashboard Cite

Rede de drenagem 2. Compartimentação automática da paisagem 3. Mapeamento digital de solos 4. Estereoscopia digital I. Título AGRADECIMENTOS Agradeço primeiramente à Deus pelo sustento e por sempre me surpreender com sua bondade. Sem isso seria impossível caminhar. Agradeço imensamente à minha família, pois sempre me apoiaram e deram suporte em todos os momentos desse caminho. Agradeço em especial ao meu Pai, Nilton. Todo o seu esforço valeu a pena. Reconheço que se não fossem os seus finais de semana perdidos eu não estaria aqui. Serei eternamente grato e sei que está orgulhoso. Agradeço à minha mãe pelo amor e carinho. A condição de fazer esse curso era você estar bem e hoje está curada. O tempo passou mas a saudade ainda aperta. Obrigado por me incentivar sempre. Agradeço à meu irmão Luccas. Nunca se contente com pouco, seja o melhor que puder em tudo. Você vai mais longe do que pode imaginar. Agradeço à Ana, minha noiva, meu amor. Você deixou a caminhada mais leve e mais bela. Obrigado pelo apoio e compreensão nos tempos difíceis. Você é uma mulher incrível! Te amo! Agradeço à minha segunda família, Nivaldo e Hilda. Obrigado por me tratarem como filho e não pouparem esforços para ajudar quando preciso. Vocês também tornaram a caminhada muito mais facil. Viva o vinho! Agradeço aos novos irmãosTaís, Pedro e Sara. A caminhada está só começando. Agradeço aos amigos e a Igreja no Jupiá.

show abstract

Digital mapping of soil properties using multiple machine learning in a semi-arid region, central Iran

Cited by 230 publications

References 54 publications

Systematic comparison of five machine-learning methods in classification and interpolation of soil particle size fractions using different transformed data

Systematic comparison of five machine-learning methods in classification and interpolation of soil particle size fractions using different transformed data

Predicting and Mapping of Soil Organic Carbon Using Machine Learning Algorithms in Northern Iran

Compartimentação da paisagem via relevo e rede de drenagem e sua relação com atributos e classes de solos

Contact Info

Product

Resources

About