Global Mapping of Soil Saturated Hydraulic Conductivity Combining Legacy Data, Spatial Covariates and Machine Learning

2021 ◽  
Author(s):  
Surya Gupta ◽  
Peter Lehmann ◽  
Andreas Papritz ◽  
Tomislav Hengl ◽  
Sara Bonetti ◽  
...  
Keyword(s):  
Machine Learning ◽  
Hydraulic Conductivity ◽  
Cross Validation ◽  
Arable Land ◽  
Remotely Sensed ◽  
Saturated Hydraulic Conductivity ◽  
Pedotransfer Functions ◽  
Data Set ◽  
Global Database ◽  
Spatially Distributed

<p>Saturated soil hydraulic conductivity (Ksat) is a key parameter in many hydrological and climatic modeling applications, as it controls the partitioning between precipitation, infiltration and runoff. Values of Ksat are often deduced from Pedotransfer Functions (PTFs) using maps of soil attributes. To circumvent inherent limitations of present PTFs (heavy reliance of arable land measurements, ignoring soil structure, and geographic bias to temperate regions), we propose a new global Ksat map at 1–km resolution by harnessing technological advances in machine learning and availability of remotely sensed surrogate information (terrain, climate and vegetation). We compiled a comprehensive Ksat data set with 13,258 data geo-referenced points from literature and other sources. The data were standardized and quality-checked in order to provide a global database of soil saturated hydraulic conductivity (SoilKsatDB). The SoilKsatDB was then applied to develop a Covariate-based GeoTransfer Function (CoGTF) model for predicting spatially distributed Ksat values using remotely sensed information on various environmental covariates. The model accuracy assessment based on spatial cross-validation shows a concordance correlation coefficient (CCC) of 0.16 and a root meansquare error (RMSE) of 1.18 for log10 Ksat values in cm/day (CCC=0.79 and RMSE=0.72 for non spatial cross-validation). The generated maps of Ksat represent spatial patterns of soil formation processes more distinctly than previous global maps of Ksat based on soil texture information and bulk density. The validation indicates that Ksat could be modeled without bias using CoGTFs that harness spatially distributed surface and climate attributes, compared to soil information based PTFs. The relatively poor performance of all models in the validation (low CCC and high RMSE) highlights the need for the collection of additional Ksat values to train the model for regions with sparse data.</p>

scholarly journals SoilKsatDB: global database of soil saturated hydraulic conductivity measurements for geoscience applications

2021 ◽  
Vol 13 (4) ◽  
pp. 1593-1612
Author(s):  
Surya Gupta ◽  
Tomislav Hengl ◽  
Peter Lehmann ◽  
Sara Bonetti ◽  
Dani Or
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Hydraulic Conductivity ◽  
Soil Properties ◽  
Bulk Density ◽  
Learning Algorithm ◽  
Field Measurements ◽  
Saturated Hydraulic Conductivity ◽  
Pedotransfer Functions ◽  
Global Database

Abstract. The saturated soil hydraulic conductivity (Ksat) is a key parameter in many hydrological and climate models. Ksat values are primarily determined from basic soil properties and may vary over several orders of magnitude. Despite the availability of Ksat datasets in the literature, significant efforts are required to combine the data before they can be used for specific applications. In this work, a total of 13 258 Ksat measurements from 1908 sites were assembled from the published literature and other sources, standardized (i.e., units made identical), and quality checked in order to obtain a global database of soil saturated hydraulic conductivity (SoilKsatDB). The SoilKsatDB covers most regions across the globe, with the highest number of Ksat measurements from North America, followed by Europe, Asia, South America, Africa, and Australia. In addition to Ksat, other soil variables such as soil texture (11 584 measurements), bulk density (11 262 measurements), soil organic carbon (9787 measurements), moisture content at field capacity (7382), and wilting point (7411) are also included in the dataset. To show an application of SoilKsatDB, we derived Ksat pedotransfer functions (PTFs) for temperate regions and laboratory-based soil properties (sand and clay content, bulk density). Accurate models can be fitted using a random forest machine learning algorithm (best concordance correlation coefficient (CCC) equal to 0.74 and 0.72 for temperate area and laboratory measurements, respectively). However, when these Ksat PTFs are applied to soil samples obtained from tropical climates and field measurements, respectively, the model performance is significantly lower (CCC = 0.49 for tropical and CCC = 0.10 for field measurements). These results indicate that there are significant differences between Ksat data collected in temperate and tropical regions and Ksat measured in the laboratory or field. The SoilKsatDB dataset is available at https://doi.org/10.5281/zenodo.3752721 (Gupta et al., 2020) and the code used to extract the data from the literature and the applied random forest machine learning approach are publicly available under an open data license.

scholarly journals SoilKsatDB: global soil saturated hydraulic conductivity measurements for geoscience applications

2020 ◽  
Author(s):  
Surya Gupta ◽  
Tomislav Hengl ◽  
Peter Lehmann ◽  
Sara Bonetti ◽  
Dani Or
Keyword(s):  
Hydraulic Conductivity ◽  
Bulk Density ◽  
Regional Scale ◽  
Field Measurements ◽  
Soil Samples ◽  
Saturated Hydraulic Conductivity ◽  
Pedotransfer Functions ◽  
Tropical Regions ◽  
Global Database ◽  
Field Samples

Abstract. Saturated soil hydraulic conductivity (Ksat) is a key parameter in many hydrological and climatic modeling applications, as it controls the partitioning between precipitation, infiltration and runoff. Ksat values are primarily determined from soil textural properties and soil forming processes, and may vary over several orders of magnitude. Despite availability of Ksat datasets at catchment or regional scale, significant efforts are required to import and bind the data before it could be used for modeling. In this work, a total of 1,910 sites with 13,267 Ksat measurements were assembled from published literature and other sources, standardized, and quality-checked in order to provide a global database of soil saturated hydraulic conductivity (SoilKsatDB). The SoilKsatDB covers most global regions, with the highest data density from the USA, followed by Europe, Asia, South America, Africa, and Australia. In addition to Ksat, other soil variables such as soil texture (11,667 measurements), bulk density (11,151 measurements), soil organic carbon (9,787 measurements), field capacity (7,389) and wilting point (7,418) are also included in the dataset. The results of using the SoilKsatDB to fit Ksat pedotransfer functions (PTFs) for temperate climatic regions and laboratory based soil samples based on soil properties (sand and clay content, bulk density) show that reasonably accurate models can be fitted using Random Forest (best CCC = 0.70 and CCC = 0.73 for temperate and lab based measurements, respectively). However when temperate and laboratory based Ksat PTFs are applied to soil samples from tropical climates and field measurements, respectively, the model performance is significantly lower (CCC = 0.51 for tropical and CCC = 0.13 for field samples). PTFs derived for temperate soils and laboratory measurements might not be suitable for estimating Ksat for tropical regions or field measurements, respectively. The SoilKsatDB dataset is available at https://doi.org/10.5281/zenodo.3752721 and the code used to produce the compilation is publicly available under an open data license.

scholarly journals An improved drip infiltrometer measuring the near-saturated hydraulic conductivity: Pedotransfer development and macropore transport

2020 ◽  
Author(s):  
Bo Vangsø Iversen ◽  
Michael Koppelgaard ◽  
Ali M. Kotlar
Keyword(s):  
Hydraulic Conductivity ◽  
Root Zone ◽  
Soil Column ◽  
Saturated Hydraulic Conductivity ◽  
Pedotransfer Functions ◽  
Ceramic Plate ◽  
Undisturbed Soil ◽  
Flow Rates ◽  
Spatially Distributed ◽  
High Level

<p>The near-saturated hydraulic conductivity is an important parameter in relation to the analysis of heterogeneous transport in the soil macropore system. To a high degree, leaching of phosphorus out of the root zone takes place in the macropores either in a dissolved form or as phosphorus bound to colloids. In this work, a newly constructed and improved drip infiltrometer (DIM) is presented being able to measure the unsaturated hydraulic conductivity in the near-saturated range (i.e. in the range of matric potentials between -0.1 and 3 -kPa) on undisturbed soil columns (20 cm by 20 cm). The DIM is a modified version of the classical multistep system establishing gravity flow at decreasing flow rates. The procedure is that the soil column is placed on top of a ceramic plate. Five tensiometers measure the change in the matric potential a different flow rates applied by a drip-irrigation device mounted on the top of the column. By applying a certain inflow at the top and suction at the bottom of the sample, a steady state flow is established based on tensiometer readings showing a constant gradient along the soil sample. This allows the determination of the near-saturated hydraulic conductivity by applying Darcy’s equation. Compared to an earlier version of the infiltrometer, the instrument has been improved in several ways. This involves a high level of automation of the computer program controlling the analysis making it possible to setup a number of settings and constrains in order to optimize the analysis. Examples are given for newly developed pedotransfer functions predicting the saturated and near-saturated hydraulic conductivity. Results were used to model water transport in the vadose zone spatially distributed over Denmark using variation in the hydraulic properties as well as spatially distributed metrological data. Models results ended up with a map pointing out risk areas of macropore transport in relation to the leaching of phosphorus.</p>

scholarly journals Saturated Hydraulic Conductivity Estimation Using Artificial Neural Networks

Water ◽  
2021 ◽  
Vol 13 (5) ◽  
pp. 705
Author(s):  
Josué Trejo-Alonso ◽  
Carlos Fuentes ◽  
Carlos Chávez ◽  
Antonio Quevedo ◽  
Alfonso Gutierrez-Lopez ◽  
...  
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Hydraulic Conductivity ◽  
Input Data ◽  
Logistic Function ◽  
Saturated Hydraulic Conductivity ◽  
Irrigation District ◽  
Pedotransfer Functions ◽  
Back Propagation Algorithm ◽  
Artificial Neural

In the present work, we construct several artificial neural networks (varying the input data) to calculate the saturated hydraulic conductivity (KS) using a database with 900 measured samples obtained from the Irrigation District 023, in San Juan del Rio, Queretaro, Mexico. All of them were constructed using two hidden layers, a back-propagation algorithm for the learning process, and a logistic function as a nonlinear transfer function. In order to explore different arrays for neurons into hidden layers, we performed the bootstrap technique for each neural network and selected the one with the least Root Mean Square Error (RMSE) value. We also compared these results with pedotransfer functions and another neural networks from the literature. The results show that our artificial neural networks obtained from 0.0459 to 0.0413 in the RMSE measurement, and 0.9725 to 0.9780 for R2, which are in good agreement with other works. We also found that reducing the amount of the input data offered us better results.

Evaluation of Pedotransfer Functions for Predicting Saturated Hydraulic Conductivity for U.S. Soils

2009 ◽  
Author(s):  
Ahmed M Abdelbaki ◽  
Mohamed A Youssef ◽  
Esmail M. F Naguib ◽  
Mohamed E Kiwan ◽  
Emad I El-giddawy
Keyword(s):  
Hydraulic Conductivity ◽  
Saturated Hydraulic Conductivity ◽  
Pedotransfer Functions

scholarly journals Classification of Brain Tumors from MRI Images Using a Convolutional Neural Network

2020 ◽  
Vol 10 (6) ◽  
pp. 1999 ◽  
Author(s):  
Milica M. Badža ◽  
Marko Č. Barjaktarović
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Brain Tumors ◽  
Convolutional Neural Network ◽  
Cross Validation ◽  
Magnetic Resonance Images ◽  
Generalization Capability ◽  
Data Set ◽  
Fold Cross Validation

The classification of brain tumors is performed by biopsy, which is not usually conducted before definitive brain surgery. The improvement of technology and machine learning can help radiologists in tumor diagnostics without invasive measures. A machine-learning algorithm that has achieved substantial results in image segmentation and classification is the convolutional neural network (CNN). We present a new CNN architecture for brain tumor classification of three tumor types. The developed network is simpler than already-existing pre-trained networks, and it was tested on T1-weighted contrast-enhanced magnetic resonance images. The performance of the network was evaluated using four approaches: combinations of two 10-fold cross-validation methods and two databases. The generalization capability of the network was tested with one of the 10-fold methods, subject-wise cross-validation, and the improvement was tested by using an augmented image database. The best result for the 10-fold cross-validation method was obtained for the record-wise cross-validation for the augmented data set, and, in that case, the accuracy was 96.56%. With good generalization capability and good execution speed, the new developed CNN architecture could be used as an effective decision-support tool for radiologists in medical diagnostics.

scholarly journals Dynamism of Landscape Transformation in Ibiono-Ibom, Akwa-Ibom State, Nigeria

2021 ◽  
Vol 14 (1) ◽  
pp. 19-38
Author(s):  
Orimoogunje Oluwagbenga O. Isaac ◽  
Adeleke Benjamin Olufemi ◽  
Dada Emmanuel ◽  
Shote Adebola Adekunle ◽  
Eudoxie-Okafor Aniefiok Nene ◽  
...  
Keyword(s):  
Agricultural Land ◽  
Arable Land ◽  
Agricultural Practices ◽  
Household Survey ◽  
Biodiversity Loss ◽  
Remotely Sensed ◽  
Structured Interviews ◽  
Remotely Sensed Data ◽  
Data Set ◽  
Landscape Transformation

Abstract Studies have shown that information on landscape transformation is an important benchmark data set because of its value as an environmental change indicator. Therefore, dynamism of landscape transformation over a 34-year period are analysed for a case study in Ibiono-Ibom, Akwa-Ibom State, Nigeria. The study adopted a mixed method consisting of remote sensing and GIS-based analysis, and semi-structured interviews covering 400 households while factors contributing to landscape structures and changes are studied. The results point out three main driving factors responsible for the landscape transformation in the study area: agricultural practices which lead to intensification of forest resources, riparian vegetation, vegetated wetlands and non-vegetated wetlands; urbanization which modifies the structure and morphology of the landscape, and finally, population growth directly related to massive infrastructural development which encroached on all other land spaces. GIS-based analysis of remotely-sensed data showed that built-up area had increased by 7535.2 ha between 1986 and 2020; shrub and arable land by 1343.9 ha and light forest decreased by 4998.3 ha. While bare-land reduced by 1522.1 ha; vegetated wetland reduced by 1092 ha; water body coverage reduced by 168 ha and non-vegetated wetland size also reduced by 2029.4 ha. Analysis of household survey results revealed that the perceptions of respondents validate the observed patterns during the remotely-sensed data analysis phase of the research, with 54 % (n=400) of respondents reporting a decline in agricultural land use, and 19.3 % (n=400) observing a decline in forest areas in the study area. Furthermore, agricultural intensification, urban development, timber exploitation, firewood collection and increase in settlements were identified as the proximate drivers of these observed landscape transformation dynamics in the study area. The study concluded that the variation in landscape transformation of the study area are clear indication of the extent of biodiversity loss and ecosystem degradation in the study area.

Sign in / Sign up

Export Citation Format

Share Document