A multi-dimensional Sentinel-based Soil Monitoring Scheme (S2MoS) for soil clay content estimation

2020 ◽  
Author(s):  
Nikolaos Tziolas ◽  
Nikolaos Tsakiridis ◽  
Eyal Ben Dor ◽  
John Theocharis ◽  
George Zalidis
Keyword(s):  
Machine Learning ◽  
Open Access ◽  
Clay Content ◽  
Bare Soil ◽  
Google Earth ◽  
Machine Learning Algorithms ◽  
Soil Monitoring ◽  
Monitoring Scheme ◽  
Soil Clay ◽  
Soil Clay Content

<p>Earth Observation (EO) has an immense potential as an enabling tool for mapping the spatial variation of the topsoil layer. Additionally, machine learning based algorithms deployed on cloud computing infrastructures have a great potential to revolutionize the processing of EO data. This paper aims to present a multi-dimensional Sentinel-based Soil Monitoring Scheme (S2MoS) based on open-access Copernicus Sentinel data and the Google Earth Engine platform to map soil properties. Building on key results from existing data mining approaches to extract bare soil reflectance values the current study presents i) preliminary insights on the synergistic use of open access SAR and optical images obtained from Sentinel-1 and Sentinel-2 sensors; and ii) evaluate the efficiency of machine learning algorithms to predict soil attributes based on multi-temporal analysis. In that regard, this study evaluated, based on Sentinel images extending over a 3 years period (2017-2019), the performance of two state of the art machine learning approaches, namely random forest and neural networks. Spatial thresholds values of 0.25 and 0.075 for Normalized Difference Vegetation Index and Normalized Burn Ratio 2 indices respectively were applied to mask bare soil pixels. In this study, we used 5000 soil data belonging to cropland land use from the European LUCAS topsoil database. We calibrated the models based on 4000 soil samples and then validated this approach with the rest 1000 samples  predict soil clay content. A higher prediction performance (R<sup>2</sup>=0.53) was achieved by the inclusion of both types (SAR and optical) of observations using the neural network model, demonstrating an improvement of about 5% in overall accuracy compared to the R<sup>2</sup> using the multi-year median optical composite.</p>

scholarly journals Employing a Multi-Input Deep Convolutional Neural Network to Derive Soil Clay Content from a Synergy of Multi-Temporal Optical and Radar Imagery Data

2020 ◽  
Vol 12 (9) ◽  
pp. 1389 ◽  
Author(s):  
Nikolaos Tziolas ◽  
Nikolaos Tsakiridis ◽  
Eyal Ben-Dor ◽  
John Theocharis ◽  
George Zalidis
Keyword(s):  
Neural Network ◽  
Open Access ◽  
Convolutional Neural Network ◽  
Near Infrared ◽  
Clay Content ◽  
Google Earth ◽  
Spectral Information ◽  
Radar Images ◽  
Soil Clay ◽  
Soil Clay Content

Earth observation (EO) has an immense potential as being an enabling tool for mapping spatial characteristics of the topsoil layer. Recently, deep learning based algorithms and cloud computing infrastructure have become available with a great potential to revolutionize the processing of EO data. This paper aims to present a novel EO-based soil monitoring approach leveraging open-access Copernicus Sentinel data and Google Earth Engine platform. Building on key results from existing data mining approaches to extract bare soil reflectance values the current study delivers valuable insights on the synergistic use of open access optical and radar images. The proposed framework is driven by the need to eliminate the influence of ambient factors and evaluate the efficiency of a convolutional neural network (CNN) to effectively combine the complimentary information contained in the pool of both optical and radar spectral information and those form auxiliary geographical coordinates mainly for soil. We developed and calibrated our multi-input CNN model based on soil samples (calibration = 80% and validation 20%) of the LUCAS database and then applied this approach to predict soil clay content. A promising prediction performance (R2 = 0.60, ratio of performance to the interquartile range (RPIQ) = 2.02, n = 6136) was achieved by the inclusion of both types (synthetic aperture radar (SAR) and laboratory visible near infrared–short wave infrared (VNIR-SWIR) multispectral) of observations using the CNN model, demonstrating an improvement of more than 5.5% in RMSE using the multi-year median optical composite and current state-of-the-art non linear machine learning methods such as random forest (RF; R2 = 0.55, RPIQ = 1.91, n = 6136) and artificial neural network (ANN; R2 = 0.44, RPIQ = 1.71, n = 6136). Moreover, we examined post-hoc techniques to interpret the CNN model and thus acquire an understanding of the relationships between spectral information and the soil target identified by the model. Looking to the future, the proposed approach can be adopted on the forthcoming hyperspectral orbital sensors to expand the current capabilities of the EO component by estimating more soil attributes with higher predictive performance.

Mean spectral reflectance from bare soil pixels along a Landsat-TM time series to increase both the prediction accuracy of soil clay content and mapping coverage

Geoderma ◽  
2021 ◽  
Vol 388 ◽  
pp. 114864
Author(s):  
Anis Gasmi ◽  
Cécile Gomez ◽  
Philippe Lagacherie ◽  
Hédi Zouari ◽  
Ahmed Laamrani ◽  
...  
Keyword(s):  
Time Series ◽  
Spectral Reflectance ◽  
Prediction Accuracy ◽  
Clay Content ◽  
Bare Soil ◽  
Landsat Tm ◽  
Soil Clay ◽  
Soil Clay Content

Litter as a filter of emergence for herbaceous seedlings and sporophytes in central Amazonia

2012 ◽  
Vol 28 (5) ◽  
pp. 445-452 ◽  
Author(s):  
Flávio Rogério de Oliveira Rodrigues ◽  
Flávia Regina Capellotto Costa
Keyword(s):  
Clay Content ◽  
Dispersal Limitation ◽  
Distribution Patterns ◽  
Bare Soil ◽  
Wet Season ◽  
Litter Addition ◽  
Litter Depth ◽  
Central Amazonia ◽  
Soil Clay ◽  
Soil Clay Content

Abstract:We conducted a study in 30 plots distributed uniformly in an area of 25 km2 at Ducke Reserve, Manaus, to test the hypothesis that understorey herb richness and abundance are mediated by litter, manipulating experimentally the amount of litter in the field. Over 10 mo, we followed the emergence of herbaceous seedlings and sporophytes in control, litter-addition and litter-exclusion treatments, covering an area of 1.2 m2 per plot in each treatment. We also assessed the relationship between topography and litter depth and frequency of bare-soil patches; and the influence of density of reproductive individuals on the emergence of herbs. Litter depth decreased, and the frequency of bare-soil patches increased with terrain slope in the wet season, but were not related with the soil clay content. Neither was related to the topography in the dry season. Emergence of pteridophytes was four times higher in the litter-exclusion treatment (3.7 ± 1.1 individuals m−2) than in the litter-addition treatment (0.9 ± 0.28 indiv. m−2) and increased with soil clay content. Seedlings from monocot herbs emerged twice more frequently in the litter exclusion (0.71 ± 0.25 indiv. m−2) than in the litter-addition treatment (0.33 ± 0.11 indiv. m−2), and also more in sites with high density of fruiting plants. The results are consistent with the hypothesis that regeneration of herbs with very small propagules is strongly affected by the physical barrier imposed by litter. Given that litter is shallower on slopes during the wet season, this creates a pattern of higher density and richness of pteridophytes in these areas. Monocot herbs, although also limited by litter, were more highly limited by availability of propagules, and their distribution patterns are at least in part explained by dispersal limitation. We conclude that litter is an important causal factor behind the herb distribution patterns along topographical gradients.

scholarly journals Exploratory Analysis of Driving Force of Wildfires in Australia: An Application of Machine Learning within Google Earth Engine

2020 ◽  
Vol 13 (1) ◽  
pp. 10
Author(s):  
Andrea Sulova ◽  
Jamal Jokar Arsanjani
Keyword(s):  
Climate Change ◽  
Machine Learning ◽  
Random Forest ◽  
Google Earth ◽  
Summer Season ◽  
Driving Factors ◽  
Machine Learning Algorithms ◽  
Classification And Regression Tree ◽  
Training Dataset ◽  
Google Earth Engine

Recent studies have suggested that due to climate change, the number of wildfires across the globe have been increasing and continue to grow even more. The recent massive wildfires, which hit Australia during the 2019–2020 summer season, raised questions to what extent the risk of wildfires can be linked to various climate, environmental, topographical, and social factors and how to predict fire occurrences to take preventive measures. Hence, the main objective of this study was to develop an automatized and cloud-based workflow for generating a training dataset of fire events at a continental level using freely available remote sensing data with a reasonable computational expense for injecting into machine learning models. As a result, a data-driven model was set up in Google Earth Engine platform, which is publicly accessible and open for further adjustments. The training dataset was applied to different machine learning algorithms, i.e., Random Forest, Naïve Bayes, and Classification and Regression Tree. The findings show that Random Forest outperformed other algorithms and hence it was used further to explore the driving factors using variable importance analysis. The study indicates the probability of fire occurrences across Australia as well as identifies the potential driving factors of Australian wildfires for the 2019–2020 summer season. The methodical approach and achieved results and drawn conclusions can be of great importance to policymakers, environmentalists, and climate change researchers, among others.

scholarly journals Remaining phosphorus and sodium fluoride pH in soils with different clay contents and clay mineralogies

2004 ◽  
Vol 39 (3) ◽  
pp. 241-246 ◽  
Author(s):  
Marcelo Eduardo Alves ◽  
Arquimedes Lavorenti
Keyword(s):  
Buffer Capacity ◽  
Ammonium Oxalate ◽  
Clay Mineralogy ◽  
Clay Content ◽  
Response Curves ◽  
Solution Ph ◽  
Ph Values ◽  
Soil Clay ◽  
Soil Clay Content ◽  
Discriminatory Capacity

The remaining phosphorus (Prem) has been used for estimating the phosphorus buffer capacity (PBC) of soils of some Brazilian regions. Furthermore, the remaining phosphorus can also be used for estimating P, S and Zn soil critical levels determined with PBC-sensible extractants and for defining P and S levels to be used not only in P and S adsorption studies but also for the establishment of P and S response curves. The objective of this work was to evaluate the effects of soil clay content and clay mineralogy on Prem and its relationship with pH values measured in saturated NaF solution (pH NaF). Ammonium-oxalate-extractable aluminum exerts the major impacts on both Prem and pH NaF, which, in turn, are less dependent on soil clay content. Although Prem and pH NaF have consistent correlation, the former has a soil-PBC discriminatory capacity much greater than pH NaF.

scholarly journals P-sorption and desorption in Savanna Brazilian soils as a support for phosphorus fertilizer management

2013 ◽  
Vol 37 (6) ◽  
pp. 521-530 ◽  
Author(s):  
Flávio Araújo Pinto ◽  
Edicarlos Damacena de Souza ◽  
Helder Barbosa Paulino ◽  
Nilton Curi ◽  
Marco Aurélio Carbone Carneiro
Keyword(s):  
Clay Content ◽  
Soil Characteristics ◽  
Fertilizer Management ◽  
Phosphorus Fertilizer ◽  
Phosphorus Adsorption ◽  
P Sorption ◽  
Brazilian Soils ◽  
Soil Clay ◽  
Sand Base ◽  
Soil Clay Content

Phosphorus (P) sorption by soils is a phenomenon that varies depending on soil characteristics, influencing its intensity and magnitude, which makes it a source or drain of P. The objective of this study was to determine the Maximum Phosphorus Adsorption Capacity (MPAC) and desorption of P from soils under native Savanna Brazilian and verify the correlation between MPAC and P Capacity Factor (PCF) with the chemical and physical properties of these soils. The study was conducted in seven soils under native Savannas. The Langmuir isotherms were adjusted from the values obtained in sorption assays, being evaluated the MPAC, the energy adsorption (EA) and PCF, which was calculated according to the levels of P-adsorbed and P-sorbed. Values of MPAC were classified as high in most soils, ranging from 283 up to 2635 mg kg-1 of P in the soil and were correlated with soil organic matter, clay, silt, sand, base saturation and pH. The PCF was higher in soils where the MPAC was also higher. The use of only one attribute of soil (clay content) as a criterion for the recommendation of phosphated fertilization, as routinely done, is susceptible to errors, needing the use of more attributes for a more accurate recommendation, as a function of the complexity of the interactions involved in the process.

Surface soil clay content mapping at large scales using multispectral (VNIR–SWIR) ASTER data

2018 ◽  
Vol 40 (4) ◽  
pp. 1506-1533
Author(s):  
Anis Gasmi ◽  
Cécile Gomez ◽  
Philippe Lagacherie ◽  
Hédi Zouari
Keyword(s):  
Surface Soil ◽  
Clay Content ◽  
Aster Data ◽  
Soil Clay ◽  
Soil Clay Content

scholarly journals Factors Affecting Microbial Formation of Nitrate-Nitrogen in Soil and Their Effects on Fertilizer Nitrogen Use Efficiency

2001 ◽  
Vol 1 ◽  
pp. 122-129 ◽  
Author(s):  
Alan Olness ◽  
Dian Lopez ◽  
David Archer ◽  
Jason Cordes ◽  
Colin Sweeney ◽  
...  
Keyword(s):  
Organic Matter ◽  
Decision Aid ◽  
Pore Space ◽  
Organic Matter Content ◽  
Clay Content ◽  
Matter Content ◽  
N Fertilizer ◽  
Nitrate N ◽  
Soil Clay ◽  
Soil Clay Content

Mineralization of soil organic matter is governed by predictable factors with nitrate-N as the end product. Crop production interrupts the natural balance, accelerates mineralization of N, and elevates levels of nitrate-N in soil. Six factors determine nitrate-N levels in soils: soil clay content, bulk density, organic matter content, pH, temperature, and rainfall. Maximal rates of N mineralization require an optimal level of air-filled pore space. Optimal air-filled pore space depends on soil clay content, soil organic matter content, soil bulk density, and rainfall. Pore space is partitioned into water- and air-filled space. A maximal rate of nitrate formation occurs at a pH of 6.7 and rather modest mineralization rates occur at pH 5.0 and 8.0. Predictions of the soil nitrate-N concentrations with a relative precision of 1 to 4 μg N g–1of soil were obtained with a computerized N fertilizer decision aid. Grain yields obtained using the N fertilizer decision aid were not measurably different from those using adjacent farmer practices, but N fertilizer use was reduced by >10%. Predicting mineralization in this manner allows optimal N applications to be determined for site-specific soil and weather conditions.

scholarly journals Integration of Machine Learning and Open Access Geospatial Data for Land Cover Mapping

2019 ◽  
Vol 11 (16) ◽  
pp. 1907 ◽  
Author(s):  
Mohammad Mardani ◽  
Hossein Mardani ◽  
Lorenzo De Simone ◽  
Samuel Varas ◽  
Naoki Kita ◽  
...  
Keyword(s):  
Machine Learning ◽  
Developing Countries ◽  
Open Access ◽  
Land Cover ◽  
Satellite Images ◽  
National Level ◽  
Google Earth ◽  
Support Vector ◽  
Land Cover Mapping ◽  
Sustainable Food

In-time and accurate monitoring of land cover and land use are essential tools for countries to achieve sustainable food production. However, many developing countries are struggling to efficiently monitor land resources due to the lack of financial support and limited access to adequate technology. This study aims at offering a solution to fill in such a gap in developing countries, by developing a land cover solution that is free of costs. A fully automated framework for land cover mapping was developed using 10-m resolution open access satellite images and machine learning (ML) techniques for the African country of Lesotho. Sentinel-2 satellite images were accessed through Google Earth Engine (GEE) for initial processing and feature extraction at a national level. Also, Food and Agriculture Organization’s land cover of Lesotho (FAO LCL) data were used to train a support vector machine (SVM) and bagged trees (BT) classifiers. SVM successfully classified urban and agricultural lands with 62 and 67% accuracy, respectively. Also, BT could classify the two categories with 81 and 65% accuracy, correspondingly. The trained models could provide precise LC maps in minutes or hours. they can also be utilized as a viable solution for developing countries as an alternative to traditional geographic information system (GIS) methods, which are often labor intensive, require acquisition of very high-resolution commercial satellite imagery, time consuming and call for high budgets.

scholarly journals Object-Oriented LULC Classification in Google Earth Engine Combining SNIC, GLCM, and Machine Learning Algorithms

2020 ◽  
Vol 12 (22) ◽  
pp. 3776
Author(s):  
Andrea Tassi ◽  
Marco Vizzari
Keyword(s):  
Machine Learning ◽  
Central Italy ◽  
Object Oriented ◽  
Google Earth ◽  
Machine Learning Algorithms ◽  
Support Vector ◽  
Landsat 8 ◽  
Good Reliability ◽  
Google Earth Engine ◽  
Occurrence Matrix

Google Earth Engine (GEE) is a versatile cloud platform in which pixel-based (PB) and object-oriented (OO) Land Use–Land Cover (LULC) classification approaches can be implemented, thanks to the availability of the many state-of-art functions comprising various Machine Learning (ML) algorithms. OO approaches, including both object segmentation and object textural analysis, are still not common in the GEE environment, probably due to the difficulties existing in concatenating the proper functions, and in tuning the various parameters to overcome the GEE computational limits. In this context, this work is aimed at developing and testing an OO classification approach combining the Simple Non-Iterative Clustering (SNIC) algorithm to identify spatial clusters, the Gray-Level Co-occurrence Matrix (GLCM) to calculate cluster textural indices, and two ML algorithms (Random Forest (RF) or Support Vector Machine (SVM)) to perform the final classification. A Principal Components Analysis (PCA) is applied to the main seven GLCM indices to synthesize in one band the textural information used for the OO classification. The proposed approach is implemented in a user-friendly, freely available GEE code useful to perform the OO classification, tuning various parameters (e.g., choose the input bands, select the classification algorithm, test various segmentation scales) and compare it with a PB approach. The accuracy of OO and PB classifications can be assessed both visually and through two confusion matrices that can be used to calculate the relevant statistics (producer’s, user’s, overall accuracy (OA)). The proposed methodology was broadly tested in a 154 km2 study area, located in the Lake Trasimeno area (central Italy), using Landsat 8 (L8), Sentinel 2 (S2), and PlanetScope (PS) data. The area was selected considering its complex LULC mosaic mainly composed of artificial surfaces, annual and permanent crops, small lakes, and wooded areas. In the study area, the various tests produced interesting results on the different datasets (OA: PB RF (L8 = 72.7%, S2 = 82%, PS = 74.2), PB SVM (L8 = 79.1%, S2 = 80.2%, PS = 74.8%), OO RF (L8 = 64%, S2 = 89.3%, PS = 77.9), OO SVM (L8 = 70.4, S2 = 86.9%, PS = 73.9)). The broad code application demonstrated very good reliability of the whole process, even though the OO classification process resulted, sometimes, too demanding on higher resolution data, considering the available computational GEE resources.

Sign in / Sign up

Export Citation Format

Share Document