The effect of covariates on Soil Organic Matter and pH variability: a digital soil mapping approach using random forest model

Annals of GIS, Journal Year: 2024, Volume and Issue: 30(2), P. 215 - 232

Published: Jan. 29, 2024

This research focuses on understanding the spatial variation of Soil Organic Matter (SOM) and pH levels in North Morocco. The study employs a comprehensive approach to enhance predictive modelling, incorporating Boruta algorithm for effective environmental covariates selection optimizing model parameters through hyperparameter optimization. Utilizing Random Forest (RF) with remote sensing indices topographic features, predicts SOM identify key contributors their variability. prediction saw significant success, notable correlation such as RVI, NDVI, TNDVI. These indices, indicative vegetation health productivity, emerged primary influencers SOM. In comparison, influence features like elevation, slope, aspect was found be less significant. Conversely, predicting challenging due minimal variability within dataset. Addressing this limitation could involve dataset expansion or alternative models low-correlated data handling. Despite RF model's limited efficacy prediction, an observable between identified, consistent prior research. Areas higher exhibited lower values, indicating relative soil acidification from organic matter decomposition. study's demonstrated potential using but enhancing is essential. Future may explore expansion, diverse sampling, testing better performance datasets. offers valuable insights advanced development enriches management practices.

Language: Английский

---

Assessing Machine Learning-Based Prediction under Different Agricultural Practices for Digital Mapping of Soil Organic Carbon and Available Phosphorus

Agriculture, Journal Year: 2022, Volume and Issue: 12(7), P. 1062 - 1062

Published: July 20, 2022

Predicting soil chemical properties such as organic carbon (SOC) and available phosphorus (Ava-P) content is critical in areas where different land uses exist. The distribution of SOC Ava-P influenced by both natural anthropogenic factors. This study aimed at (1) predicting a piedmont plain Northeast Iran using the Random Forests (RF) Cubist mathematical models hybrid (Regression Kriging), (2) comparing models’ results, (3) identifying key variables that influence spatial dynamics under agricultural practices. machine learning were trained with 201 composite surface samples 24 ancillary data, including climate (C), organism (O), topography- relief (R), parent material (P) features (S) according to SCORPAN digital mapping framework, which can predictively represent formation factors spatially. Clay, one most well-known relationship SOC, was important predictor followed open-access multispectral satellite images-based vegetation indices. had similar set effective variables. Hybrid approaches did not improve model accuracy significantly, but they reduce map uncertainty. In validation set, calculated RF algorithm normalized root mean square (NRMSE) 96.8, while an NRMSE 94.2. These values change when technique for Ava-P; however, changed just 1% SOC. management supply activities be guided maps. Produced maps scientist plays active role used identify concentrations are high need protected, uncertainty sampling required further monitoring.

Language: Английский

---

Sentinel-2 and Sentinel-1 Bare Soil Temporal Mosaics of 6-year Periods for Soil Organic Carbon Content Mapping in Central France

Remote Sensing, Journal Year: 2023, Volume and Issue: 15(9), P. 2410 - 2410

Published: May 4, 2023

Satellite-based soil organic carbon content (SOC) mapping over wide regions is generally hampered by the low sampling density and diversity of periods. Some unfavorable topsoil conditions, such as high moisture, rugosity, presence crop residues, limited amplitude SOC values area bare when a single image used, are also among influencing factors. To generate reliable map, this study addresses use Sentinel-2 (S2) temporal mosaics (S2Bsoil) 6 years jointly with moisture products (SMPs) derived from Sentinel 1 2 images, measurement data other environmental covariates digital elevation models, lithology maps airborne gamma-ray data. In study, we explore (i) dates periods that preferable to construct soils while accounting for management; (ii) which set more relevant explain variability. From four sets covariates, best contributing was selected, median along uncertainty at 90% prediction intervals were mapped 25-m resolution quantile regression forest models. The accuracy predictions assessed 10-fold cross-validation, repeated five times. models using all had model performance. Airborne thorium, slope S2 bands (e.g., 6, 7, 8, 8a) indices calcareous sedimentary rocks, “calcl”) “late winter–spring” time series most important in model. Our results indicated role neighboring topographic distances oblique geographic coordinates between remote sensing parent material. These contributed not only optimizing performance but provided information related long-range gradients spatial variability, makes sense pedological point view.

Language: Английский

---

Fine-resolution mapping of cropland topsoil pH of Southern China and its environmental application

Geoderma, Journal Year: 2024, Volume and Issue: 442, P. 116798 - 116798

Published: Feb. 1, 2024

Soil pH is one of the critical indicators soil quality. A fine resolution map urgently required to address practical issues agricultural production, environmental protection, and ecosystem functioning, which often fall short meeting demands for local applications. To fill this gap, we used data from an extensive survey 13,424 surface samples (0–0.2 m) across cropland Jiangxi Province in Southern China. Using digital mapping techniques with 46 covariates, produced a 30 m topsoil We integrate different variable selection algorithms machine learning methods. Our results indicate Random Forest covariates selected by recursive feature had best performance r 0.583 RMSE 0.41. The prediction interval coverage probability our was 0.92, indicating low estimated uncertainty. Climate identified as most predicting contribution 37.42 %, followed properties (29.09 %), management (21.86 parent material (6.22 biota (5.39 %) factors. mean 5.21, great pressure acidification region. high values were mainly distributed Northern, Western, Eastern parts region while majorly located central part. Compared past surveys 1980 s, there no significant change surveyed can provide important implications guidance decisions on heavy metal pollution remediation, precision agriculture, prevention acidification.

Language: Английский

---

Using local ensemble models and Landsat bare soil composites for large-scale soil organic carbon maps in cropland

Geoderma, Journal Year: 2024, Volume and Issue: 444, P. 116850 - 116850

Published: March 19, 2024

National soil organic carbon (SOC) maps are essential to improve greenhouse gas accounting and support climate-smart agriculture. Large-scale SOC models based on wall-to-wall information from remote sensing remain a challenge due the high diversity of natural conditions difficulty for spatial location samples. In this study, we tested if implementation local ensemble (LEM) can be used predictions Landsat-based reflectance composites (SRC) Germany. For this, divided research area into 30 times km tiles calculated generalized linear (GLM) random, nearby observations. Based GLMs, were predicted aggregated using moving window approach. The variable importance was analyzed identify dependencies in correlation between SRC SOC. final map, Random Forest (RF) model trained predictions, SRC, full set training samples agricultural inventory. results show that LEM able accuracy (R2 = 0.68; RMSE 5.6 g kg−1), compared single, global 0.52; 6.8 kg−1). spectral bands showed clear patterns throughout area. Differences explained by conditions, influencing properties. Compared widely adopted integration distance covariates such as geographical coordinates, reduce autocorrelation greater extent prediction accuracy, especially underrepresented values. presents new method integrate increase interpretability DSM models.

Language: Английский

---

Influence of Soil Texture on the Estimation of Soil Organic Carbon From Sentinel‐2 Temporal Mosaics at 34 European Sites

European Journal of Soil Science, Journal Year: 2025, Volume and Issue: 76(1)

Published: Jan. 1, 2025

ABSTRACT Multispectral imaging satellites such as Sentinel‐2 are considered a possible tool to assist in the mapping of soil organic carbon (SOC) using images bare soil. However, reported results variable. The measured reflectance surface is not only related SOC but also several other environmental and edaphic factors. Soil texture one factor that strongly affects reflectance. Depending on spatial correlation with SOC, influence may improve or hinder estimation from spectral data. This study aimed investigate these influences local models at 34 sites different pedo‐climatic zones across 10 European countries. were individual agricultural fields few close proximity. For each site, predict clay particle size fraction developed temporal mosaics images. Overall, predicting was difficult, prediction performances ratio performance deviation (RPD) > 1.5 observed 8 12 for clay, respectively. A general relationship between evident explained small part large variability we sites. Adding information additional predictors improved average, benefit varied average relative importance bands indicated red far‐red regions visible spectrum more important than prediction. opposite true region around 2200 nm, which models.

Language: Английский

---

Digital Mapping of Soil Organic Carbon Based on Machine Learning and Regression Kriging

Sensors, Journal Year: 2022, Volume and Issue: 22(22), P. 8997 - 8997

Published: Nov. 21, 2022

In the last two decades, machine learning (ML) methods have been widely used in digital soil mapping (DSM), but regression kriging (RK) model which combines advantages of ML and has rarely DSM. addition, due to limitation a single-model structure, many poor prediction accuracy undulating terrain areas. this study, we collected SOC content 115 samples hilly farming area with continuous terrain. According theory soil-forming factors pedogenesis, selected 10 topographic indices, 7 vegetation 2 indices as environmental covariates, according law geographical similarity, RK mine relationship between covariates predict content. Four ensemble models—random forest (RF), Cubist, stochastic gradient boosting (SGB), Bayesian regularized neural networks (BRNNs)—were fit trend content, simple (SK) method was interpolate residuals models, then residual were superimposed obtain result. Moreover, divided into calibration validation sets at ratio 80%, tenfold cross-validation optimal parameters model. From results four models: RF performed best set (R2c = 0.834) poorly (R2v 0.362); Cubist had good stability both 0.693 R2v 0.445); SGB 0.430 0.336); BRNN lowest 0.323 0.282). The showed that R2 models 0.718, 0.674, 0.724, 0.625, respectively. Compared without residuals, improved by 0.356, 0.229, 0.388, 0.343, conclusion, high generalization ability areas complex topography, can make full use trends spatial structural are not easy effectively improve accuracy. This provides reference for survey

Language: Английский

---

Remotely sensed inter-field variation in soil organic carbon content as influenced by the cumulative effect of conservation tillage in northeast China

Soil and Tillage Research, Journal Year: 2024, Volume and Issue: 243, P. 106170 - 106170

Published: May 30, 2024

Language: Английский

---

Assessing the potential of multi-source remote sensing data for cropland soil organic matter mapping in hilly and mountainous areas

CATENA, Journal Year: 2024, Volume and Issue: 245, P. 108312 - 108312

Published: Aug. 12, 2024

Language: Английский

---

Statistical evaluation of multiple interpolation techniques for spatial mapping of highly variable geotechnical facets of soil in natural deposition

Earth Science Informatics, Journal Year: 2023, Volume and Issue: 16(1), P. 105 - 129

Published: Jan. 5, 2023

Language: Английский

---