scholarly journals Wavelet geographically weighted regression for spectroscopic modelling of soil properties

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Yongze Song ◽  
Zefang Shen ◽  
Peng Wu ◽  
R. A. Viscarra Rossel
Keyword(s):  
Organic Carbon ◽  
Soil Properties ◽  
Geographically Weighted Regression ◽  
Near Infrared ◽  
Linear Models ◽  
Spatial Information ◽  
Clay Content ◽  
Weighted Regression ◽  
Organic Carbon Content ◽  
Least Squares Regression

AbstractSoil properties, such as organic carbon, pH and clay content, are critical indicators of ecosystem function. Visible–near infrared (vis–NIR) reflectance spectroscopy has been widely used to cost-efficiently estimate such soil properties. Multivariate modelling, such as partial least squares regression (PLSR), and machine learning are the most common methods for modelling soil properties with spectra. Often, such models do not account for the multiresolution information presented in the vis–NIR signal, or the spatial variation in the data. To address these potential shortcomings, we used wavelets to decompose the vis–NIR spectra of 226 soils from agricultural and forested regions in south-western Western Australia and developed a wavelet geographically weighted regression (WGWR) for estimating soil organic carbon content, clay content and pH. To evaluate the WGWR models, we compared them to linear models derived with multiresolution data from a wavelet decomposition (WLR) and PLSR without multiresolution information. Overall, validation of the WGWR models produced more accurate estimates of the soil properties than WLR and PLSR. Around 3.5–49.1% of the improvement in the estimates was due to the multiresolution analysis and 1.0–5.2% due to the integration of spatial information in the modelling. The WGWR improves the modelling of soil properties with spectra.

scholarly journals Estimation of effective calibration sample size using visible near infrared spectroscopy: deep learning vs machine learning

2019 ◽  
Author(s):  
Wartini Ng ◽  
Budiman Minasny ◽  
Wanderson de Sousa Mendes ◽  
José A. M. Demattê
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Soil Properties ◽  
Sample Size ◽  
Near Infrared ◽  
Model Performance ◽  
Clay Content ◽  
Least Squares Regression ◽  
Soil Attributes ◽  
Calibration Dataset

Abstract. The number of samples used in the calibration dataset affects the quality of the generated predictive models using visible, near and shortwave infrared (VIS-NIR-SWIR) spectroscopy for soil attributes. Recently, convolutional neural network (CNN) is regarded as a highly accurate model for predicting soil properties on a large database, however it has not been ascertained yet how large the sample size should be for CNN model to be effective. This paper aims at providing an estimate of how much calibration samples are needed to improve the model performance of soil properties predictions with CNN. It is hypothesized that the larger the amount of data, the more accurate is the CNN model. The performance of two commonly used machine learning models (Partial least squares regression (PLSR) and Cubist) are compared against the CNN model. A VIS-NIR-SWIR spectral library from Brazil containing 4251 unique sites, with averages of 2–3 samples per depth (a total of 12,044 samples), was divided into calibration (3188 sites) and validation (1063 sites) sets. A subset of the calibration dataset was then created to represent smaller calibration dataset ranging from 125, 300, 500, 1000, 1500, 2000, 2500 and 2700 unique sites, or equivalent to sample size approximately 350, 840, 1400, 2800, 4200, 5600, 7000, and 7650. All three models (PLSR, Cubist, and CNN models) were generated for each sample size of the unique sites for the prediction of five different soil properties, i.e. cation exchange capacity, organic matter, sand, silt and clay content. These calibration subset sampling processes and modelling were repeated ten times to provide a better representation of the model performances. Similar results were observed when the performances of both PLSR and Cubist model were compared to the CNN model where the performance of CNN outweighed the PLSR and Cubist model at sample size of 1500 and 1800 respectively. It can be recommended that deep learning is most efficient for spectral modelling for sample size above 2000. The accuracy of the PLSR and Cubist model seemed to reach a plateau above sample size of 4200 and 5000 respectively. A sensitivity analysis was performed on the CNN model to determine important wavelengths region that affected the predictions of various soil attributes.

scholarly journals A Comparative Assessment of Geostatistical, Machine Learning, and Hybrid Approaches for Mapping Topsoil Organic Carbon Content

2019 ◽  
Vol 8 (4) ◽  
pp. 174 ◽  
Author(s):  
Lin Chen ◽  
Chunying Ren ◽  
Lin Li ◽  
Yeqiao Wang ◽  
Bai Zhang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Organic Carbon ◽  
Geographically Weighted Regression ◽  
Ordinary Kriging ◽  
Linear Models ◽  
Weighted Regression ◽  
Hybrid Approaches ◽  
Better Than

Accurate digital soil mapping (DSM) of soil organic carbon (SOC) is still a challenging subject because of its spatial variability and dependency. This study is aimed at comparing six typical methods in three types of DSM techniques for SOC mapping in an area surrounding Changchun in Northeast China. The methods include ordinary kriging (OK) and geographically weighted regression (GWR) from geostatistics, support vector machines for regression (SVR) and artificial neural networks (ANN) from machine learning, and geographically weighted regression kriging (GWRK) and artificial neural networks kriging (ANNK) from hybrid approaches. The hybrid approaches, in particular, integrated the GWR from geostatistics and ANN from machine learning with the estimation of residuals by ordinary kriging, respectively. Environmental variables, including soil properties, climatic, topographic, and remote sensing data, were used for modeling. The mapping results of SOC content from different models were validated by independent testing data based on values of the mean error, root mean squared error and coefficient of determination. The prediction maps depicted spatial variation and patterns of SOC content of the study area. The results showed the accuracy ranking of the compared methods in decreasing order was ANNK, SVR, ANN, GWRK, OK, and GWR. Two-step hybrid approaches performed better than the corresponding individual models, and non-linear models performed better than the linear models. When considering the uncertainty and efficiency, ML and two-step approach are more suitable than geostatistics in regional landscapes with the high heterogeneity. The study concludes that ANNK is a promising approach for mapping SOC content at a local scale.

Reactions of copper with soil affecting its availability to plants. I. Effect of soil type and time

Soil Research ◽  
1980 ◽  
Vol 18 (4) ◽  
pp. 447 ◽  
Author(s):  
RF Brennan ◽  
JW Gartrell ◽  
AD Robson
Keyword(s):  
Organic Carbon ◽  
Soil Properties ◽  
Carbon Content ◽  
Soil Type ◽  
Clay Content ◽  
Organic Carbon Content ◽  
The Difference ◽  
Western Australian

The effect of moist incubation on the availability of applied copper to wheat was examined in a range of Western Australian soils. Incubating soil with copper reduced its availability relative to freshly applied copper by up to 70%. The availability of copper to wheat plants decreased with increasing time of incubation up to 120 days. The extent of the decline in availability differed among soils. The difference did not appear to be specifically related to any one of the following soil properties-pH, organic carbon content, clay content, free sesquioxide content and levels of total and extractable copper.

scholarly journals Comparison of Field and Laboratory Wet Soil Spectra in the Vis-NIR Range for Soil Organic Carbon Prediction in the Absence of Laboratory Dry Measurements

2020 ◽  
Vol 12 (18) ◽  
pp. 3082
Author(s):  
James Kobina Mensah Biney ◽  
Luboš Borůvka ◽  
Prince Chapman Agyeman ◽  
Karel Němeček ◽  
Aleš Klement
Keyword(s):  
Organic Carbon ◽  
Soil Organic Carbon ◽  
Soil Properties ◽  
Near Infrared ◽  
Rapid Determination ◽  
Principal Component Regression ◽  
Principal Component ◽  
Support Vector ◽  
Least Squares Regression ◽  
Orthogonal Signal Correction

Spectroscopy has demonstrated the ability to predict specific soil properties. Consequently, it is a promising avenue to complement the traditional methods that are costly and time-consuming. In the visible-near infrared (Vis-NIR) region, spectroscopy has been widely used for the rapid determination of organic components, especially soil organic carbon (SOC) using laboratory dry (lab-dry) measurement. However, steps such as collecting, grinding, sieving and soil drying at ambient (room) temperature and humidity for several days, which is a vital process, make the lab-dry preparation a bit slow compared to the field or laboratory wet (lab-wet) measurement. The use of soil spectra measured directly in the field or on a wet sample remains challenging due to uncontrolled soil moisture variations and other environmental conditions. However, for direct and timely prediction and mapping of soil properties, especially SOC, the field or lab-wet measurement could be an option in place of the lab-dry measurement. This study focuses on comparison of field and naturally acquired laboratory measurement of wet samples in Visible (VIS), Near-Infrared (NIR) and Vis-NIR range using several pretreatment approaches including orthogonal signal correction (OSC). The comparison was concluded with the development of validation models for SOC prediction based on partial least squares regression (PLSR) and support vector machine (SVMR). Nonetheless, for the OSC implementation, we use principal component regression (PCR) together with PLSR as SVMR is not appropriate under OSC. For SOC prediction, the field measurement was better in the VIS range with R2CV = 0.47 and RMSEPcv = 0.24, while in Vis-NIR range the lab-wet measurement was better with R2CV = 0.44 and RMSEPcv = 0.25, both using the SVMR algorithm. However, the prediction accuracy improves with the introduction of OSC on both samples. The highest prediction was obtained with the lab-wet dataset (using PLSR) in the NIR and Vis-NIR range with R2CV = 0.54/0.55 and RMSEPcv = 0.24. This result indicates that the field and, in particular, lab-wet measurements, which are not commonly used, can also be useful for SOC prediction, just as the lab-dry method, with some adjustments.

Effects of different sampling densities on geographically weighted regression kriging for predicting soil organic carbon

2017 ◽  
Vol 20 ◽  
pp. 76-91 ◽  
Author(s):  
Huichun Ye ◽  
Wenjiang Huang ◽  
Shanyu Huang ◽  
Yuanfang Huang ◽  
Shiwen Zhang ◽  
...  
Keyword(s):  
Organic Carbon ◽  
Soil Organic Carbon ◽  
Geographically Weighted Regression ◽  
Weighted Regression ◽  
Regression Kriging

scholarly journals Improvement of Tea (Camellia sinensis L.) Soil Properties by Growing Different Green Crops

2015 ◽  
Vol 12 (2) ◽  
pp. 34-38 ◽  
Author(s):  
Ashim Kumar Saha ◽  
Apu Biswas ◽  
Abdul Qayyum Khan ◽  
Md. Mohashin Farazi ◽  
Md. Habibur Rahman
Keyword(s):  
Organic Carbon ◽  
Soil Properties ◽  
Total Nitrogen ◽  
Soil Ph ◽  
Nutritional Value ◽  
Soil Health ◽  
Critical Value ◽  
Available Phosphorus ◽  
Organic Carbon Content

Long-term tea cultivation has led to degradation of the soil. Old tea soils require rehabilitation for restoring soil health. Soil rehabilitation by growing different green crops can break the chain of monoculture of tea. An experiment was conducted at The Bangladesh Tea Research Institute (BTRI) Farm during 2008-2011 to find out the efficiency of different green crops on the improvement of soil properties. Four green crops such as Guatemala, Citronella, Mimosa and Calopogonium were grown to develop the nutritional value of the degraded tea soil. Soil samples were collected and analyzed before and at the end of experiment. Soil pH was increased in all four green crops treated plots with the highest increase in Citronella treated plots (from 4.1 to 4.5). Highest content of organic carbon (1.19%) and total nitrogen (0.119%) were found in Mimosa and Calopogonium treated plots, respectively. Concentration of available phosphorus, calcium and magnesium in all green crops treated plots were above the critical values, while available potassium content was above the critical value in Guatemala, Citronella and Mimosa treated plots. Changes in soil pH and available potassium were significant, while changes in organic carbon content, total nitrogen and available calcium were insignificant. Changes in available phosphorus and magnesium were significant. The Agriculturists 2014; 12(2) 34-38

Relationships between extractable Al, selected soil properties, pH buffer capacity and lime requirement in some acidic Queensland soils

Soil Research ◽  
1992 ◽  
Vol 30 (2) ◽  
pp. 119 ◽  
Author(s):  
RL Aitken
Keyword(s):  
Organic Matter ◽  
Organic Carbon ◽  
Soil Properties ◽  
Buffer Capacity ◽  
Clay Content ◽  
Ph Buffering ◽  
Ph Buffer ◽  
Mineral Soils ◽  
Chloride Salts ◽  
Hydrolysis Of

The objectives of this study were to examine (1) interrelationships between various forms of extractable A1 and selected soil properties, (2) the contribution of extractable A1 to pH buffer capacity, and (3) investigate the use of extractable A1 to predict lime requirement. Aluminium was extracted from each of 60 Queensland soils with a range of chloride salts: 1 M KCl (AlK), 0.5 M CuCl2 (AlCu), 0.33 M LaCl3 (AlLa) and 0.01 M CaCl2 (AlCa). The amounts of A1 extracted were in the order AlCu > AlLa > Alk > AlCa. Little or no A1 was extracted by KC1 or Lac13 in soils with pHw values greater than 5.5 , whereas CuCl2 extracted some A1 irrespective of soil pH. The greater amounts of A1 extracted by CuCl2 were attributed mainly to A1 from organic matter, even though all of the soils were mineral soils (organic carbon 54.7%). Both AlCu and AlLa, were significantly (P < 0.001) correlated with organic carbon, whereas none of the extractable A1 measures was correlated with clay content. AlK and A~L, were poorly correlated to pH buffer capacity. The linear relationship between AlCu and pH buffer capacity (r2 = 0.49) obtained in this study supports the view of previous researchers that the hydrolysis of A1 adsorbed by organic matter is a source of pH buffering in soils. However, the change in CEC with pH accounted for 76% of the variation in pH buffer capacity, indicating that other mechanisms such as deprotonation of organic groups and variable charge minerals are also involved in pH buffering. The ability of CuCl2 and LaCl3extractable Al to estimate lime requirement depended on the target pH. The results suggest that lime requirements based on neutralization of AlLa would be sufficient to raise pHw to around 5.5, whereas requirements based on neutralization of AlCu substantially overestimated the actual lime requirement to pHw 5.5, but gave a reasonable estimation of the lime requirement to pHw 6 5.

Sign in / Sign up

Export Citation Format

Share Document