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

---

Digital mapping of soil organic carbon using remote sensing data: A systematic review

CATENA, Journal Year: 2023, Volume and Issue: 232, P. 107409 - 107409

Published: July 27, 2023

Language: Английский

---

A novel method for detecting soil salinity using AVIRIS-NG imaging spectroscopy and ensemble machine learning

ISPRS Journal of Photogrammetry and Remote Sensing, Journal Year: 2023, Volume and Issue: 200, P. 191 - 212

Published: May 23, 2023

Language: Английский

---

High-Precision Mapping of Soil Organic Matter Based on UAV Imagery Using Machine Learning Algorithms

Drones, Journal Year: 2023, Volume and Issue: 7(5), P. 290 - 290

Published: April 26, 2023

Soil organic matter (SOM) is a critical indicator of soil nutrient levels, and the precise mapping its spatial distribution through remote sensing essential for regulation, fertilization, scientific management protection. This information can offer decision support to agricultural departments various producers. In this paper, two new indices, NLIrededge2 GDVIrededge2, were proposed based on sensitive spectral response characteristics SOM in Northeast China. Nine parameters suitable modeling determined using competitive adaptive reweighted sampling (CARS) method, combined with spectrum reflectance, mathematical transformations vegetation so on. Then, utilizing unmanned aerial vehicle (UAV)-based multispectral images centimeter-level resolution, random forest machine learning algorithm was used construct inversion model study area. The results showed that performed best estimating (R2 = 0.91, RMSE 0.95, MBE 0.49, RPIQ 3.25) when compared other algorithms such as vector regression (SVR), elastic net, Bayesian ridge, linear regression. findings indicated negative correlation between content altitude. concluded could meet needs farmers obtain basic provide reference UAVs monitor SOM.

Language: Английский

---

National-scale mapping of soil organic carbon stock in France: New insights and lessons learned by direct and indirect approaches

Soil & Environmental Health, Journal Year: 2023, Volume and Issue: 1(4), P. 100049 - 100049

Published: Nov. 11, 2023

Soil organic carbon (SOC) plays a crucial role in soil health and global cycling, therefore accurate estimates of its spatial distribution are important for managing mitigating climate change. Digital mapping shows potential to provide high-resolution SOC across scales. To convert content density (SOCD), two inference trajectories exist predicting SOCD digital mapping: the direct approach (calculate-then-model) indirect (model-then-calculate). However, there is lack comprehensive exploration regarding differences their performance estimates, particularly regions characterized by diverse pedoclimatic conditions. bridge this knowledge gap, we evaluated approaches based on model France. Using 916 topsoils (0−20 cm) from LUCAS 2018 24 environmental covariates, random forest forward recursive feature selection were used build predictive models using approaches. The results show that, full both showed moderate (R2 = 0.28−0.32). By utilizing model, number predictors was reduced 9, enhancing 0.35) with no improvement 0.28). mean French topsoil 5.29 6.14 kg m-2 approaches, resulting stock (SOCS) 2.8 3.3 Pg, respectively. We found that clearly underestimated high (>9 m-2), while performed much better SOCD. Our findings serve as valuable reference mapping, thereby providing scientific basis maintaining health.

Language: Английский

---

Improving forest age prediction performance using ensemble learning algorithms base on satellite remote sensing data

Ecological Indicators, Journal Year: 2024, Volume and Issue: 166, P. 112327 - 112327

Published: July 16, 2024

Forest age plays a crucial role in assessing forest structure, carbon sinks, and other ecological functions. How to estimate by satellite remote sensing data has been hot research topic. This study focused on the forests of Zhejiang Province, utilizing Landsat 5 as source extract distribution information broadleaf coniferous forests. Then, feature variables were screened, was estimated using multiple linear regression model MLR, machine learning (K-nearest neighbor method KNN, support vector SVR), ensemble (adaptive boosting AdaBoost, random RF, eXtreme gradient XGBoost). After analyzing estimation results from different models, best-performing selected create spatial map Province. The shows that can better realize inversion age. optimal for is XGBoost model, with coefficient determination R2 0.832, root mean square error (RMSE) 5.823a relative (rRMSE) 21.009%. And top RF 0.800, RMSE 5.076a rRMSE 19.782%. Compared MLR best models improved 75.120% 82.500%, reduced 52.674% 47.540%, 52.703% 47.480% respectively. Additionally, analysis revealed 50% involved are texture features, indicating an important variable construction models.

Language: Английский

---

Spatial Estimation of Soil Organic Carbon Content Utilizing PlanetScope, Sentinel-2, and Sentinel-1 Data

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(17), P. 3268 - 3268

Published: Sept. 3, 2024

The accurate prediction of soil organic carbon (SOC) is important for agriculture and land management. Methods using remote sensing data are helpful estimating SOC in bare soils. To overcome the challenge predicting under vegetation cover, this study extracted spectral, radar, topographic variables from multi-temporal optical satellite images (high-resolution PlanetScope medium-resolution Sentinel-2), synthetic aperture radar (Sentinel-1), digital elevation model, respectively, to estimate content arable soils Wuling Mountain region Southwest China. These were modeled at four different spatial resolutions (3 m, 20 30 80 m) eXtreme Gradient Boosting algorithm. results showed that modeling resolution, combination multi-source data, temporal phases all influenced performance. models generally yielded better a medium (20 resolution than fine coarse (80 resolutions. PlanetScope, Sentinel-2, topography factors gave satisfactory predictions dry (R2 = 0.673, MAE 0.107%, RMSE 0.135%). addition Sentinel-1 indicators best paddy field 0.699, 0.114%, 0.148%). values R2 optimal improved by 36.0% 33.4%, compared entire area. winter played dominant role both land. This offers valuable insights into effectively properties cover various scales data.

Language: Английский

---

Preliminary Results in Innovative Solutions for Soil Carbon Estimation: Integrating Remote Sensing, Machine Learning, and Proximal Sensing Spectroscopy

Remote Sensing, Journal Year: 2023, Volume and Issue: 15(23), P. 5571 - 5571

Published: Nov. 30, 2023

This paper explores the application and advantages of remote sensing, machine learning, mid-infrared spectroscopy (MIR) as a popular proximal sensing tool in estimation soil organic carbon (SOC). It underscores practical implications benefits integrated approach combining for SOC prediction across range applications, including comprehensive health mapping credit assessment. These advanced technologies offer promising pathway, reducing costs resource utilization while improving precision estimation. We conducted comparative analysis between MIR-predicted values laboratory-measured using 36 samples. The results demonstrate strong fit (R² = 0.83), underscoring potential this approach. While acknowledging that our is based on limited sample size, these initial findings promise serve foundation future research. will be providing updates when we obtain more data. Furthermore, commercialising Australia, with aim helping farmers harness markets. Based study’s findings, coupled insights from existing literature, suggest adopting measurement could significantly benefit local economies, enhance farmers’ ability to monitor changes health, promote sustainable agricultural practices. outcomes align global climate change mitigation efforts. approach, supported by other research, offers template regions worldwide seeking similar solutions.

Language: Английский

---

Using Sentinel-1 Time Series Data for the Delineation of Management Zones

AgriEngineering, Journal Year: 2025, Volume and Issue: 7(5), P. 150 - 150

Published: May 8, 2025

The characterization of soil attribute variability often requires dense sampling grids, which can be economically unfeasible. A possible solution is to perform targeted based on previously collected data. objective this research was develop a method for mapping attributes Management Zones (MZs) delineated from Sentinel-1 radar images were used create time profiles six indices VV (vertical–vertical) and VH (vertical–horizontal) backscatter in two agricultural fields. MZs by analyzing VV/VH bands individually through approaches: (1) fuzzy k-means clustering directly applied the indices’ series (2) dimensionality reduction using deep-learning autoencoders followed clustering. best combination index MZ delineation approaches compared with four methods: conventional (single composite sample), high-density uniform grid (one sample per hectare), rectangular cells cell 5 10 hectares), random varying sizes). Leave-one-out cross-validation evaluated performance each method. Results showed that combining provided more accurate estimates, outperforming conventional, cells, In conclusion, proposed methodology presents scalability potential, as it does not require prior calibration validated types commonly found across Brazil’s regions, making suitable integration into digital platforms broader application precision agriculture.

Language: Английский

---

Continental-scale mapping of soil pH with SAR-optical fusion based on long-term earth observation data in google earth engine

Ecological Indicators, Journal Year: 2024, Volume and Issue: 165, P. 112246 - 112246

Published: June 14, 2024

Language: Английский

---