High Resolution Mapping of Soil Properties Using Remote Sensing Variables in South-Western Burkina Faso: A Comparison of Machine Learning and Multiple Linear Regression Models

Картирование свойств почвы с высоким пространственным разрешением с использованием переменных дистанционного зондирования на юго-западе Буркина-Фасо: сравнение моделей машинного обучения и множественной линейной регрессии
Michael A. Thiel, Ozias Hounkpatin, Gerald Forkuor, Gerhard Welp
2017-01-23

Burkina Fasodigital soil mappingrandom forest regressionremote sensingsoil organic carbon
Accurate and detailed spatial soil information is essential for environmental modelling, risk assessment and decision making. The use of Remote Sensing data as secondary sources of information in digital soil mapping has been found to be cost effective and less time consuming compared to traditional soil mapping approaches. But the potentials of Remote Sensing data in improving knowledge of local scale soil information in West Africa have not been fully explored. This study investigated the use of high spatial resolution satellite data (RapidEye and Landsat), terrain/climatic data and laboratory analysed soil samples to map the spatial distribution of six soil properties-sand, silt, clay, cation exchange capacity (CEC), soil organic carbon (SOC) and nitrogen-in a 580 km2 agricultural watershed in south-western Burkina Faso. Four statistical prediction models-multiple linear regression (MLR), random forest regression (RFR), support vector machine (SVM), stochastic gradient boosting (SGB)-were tested and compared. Internal validation was conducted by cross validation while the predictions were validated against an independent set of soil samples considering the modelling area and an extrapolation area. Model performance statistics revealed that the machine learning techniques performed marginally better than the MLR, with the RFR providing in most cases the highest accuracy. The inability of MLR to handle non-linear relationships between dependent and independent variables was found to be a limitation in accurately predicting soil properties at unsampled locations. Satellite data acquired during ploughing or early crop development stages (e.g. May, June) were found to be the most important spectral predictors while elevation, temperature and precipitation came up as prominent terrain/climatic variables in predicting soil properties. The results further showed that shortwave infrared and near infrared channels of Landsat8 as well as soil specific indices of redness, coloration and saturation were prominent predictors in digital soil mapping. Considering the increased availability of freely available Remote Sensing data (e.g. Landsat, SRTM, Sentinels), soil information at local and regional scales in data poor regions such as West Africa can be improved with relatively little financial and human resources.
1
Elevation, temperature, precipitation, Landsat 8 shortwave-infrared and near-infrared bands, and soil-specific redness indices were prominent predictors of soil-property distributions.
2
High-resolution RapidEye, Landsat, terrain, climatic, and soil-sample data were used to map sand, silt, clay, CEC, SOC, and nitrogen across a 580 km² agricultural watershed in Burkina Faso.
3
Multiple linear regression was limited by its inability to represent nonlinear relationships between soil properties and environmental predictors, reducing prediction accuracy at unsampled locations.
4
Random forest regression, support vector machine, and stochastic gradient boosting performed marginally better than multiple linear regression, with random forest generally achieving the highest accuracy.
5
Satellite imagery acquired during ploughing or early crop development, particularly in May and June, provided the most important spectral predictors of soil properties.

Soil properties in a 580 km² agricultural watershed in south-western Burkina Faso

Spatial prediction and mapping accuracy of sand, silt, clay, cation exchange capacity, soil organic carbon, and nitrogen using remote sensing, terrain/climatic variables, and comparative regression and machine-learning models

Publication Details
Publication Date
2017-01-23
Journal
Publisher
ISSN
Cited by
517
Access Type
Author Information
Authors
Michael A. Thiel
Ozias Hounkpatin
Gerald Forkuor
Gerhard Welp
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%