Estimation of soil properties at the field scale from satellite data: a comparison between spatial and non‐spatial techniques

Abstract

Summary A study was carried out to investigate the usefulness of multispectral and hyperspectral satellite information for the estimation of soil properties of agronomic importance such as soil texture and organic matter (SOM) in cultivated fields by comparing different estimation procedures. Images acquired from the Advanced Land Imager (ALI) and Hyperion sensors on board the EO‐1 satellite were used, in combination with ground‐sampling data from an agricultural field in central Italy, to evaluate the advantage of taking into account the spatial correlation between pixels. For this purpose, partial least squares regression (PLSR), ordinary least square (OLS) regression, regression with correlated errors (restricted maximum likelihood; REML) and ordinary kriging (OK) were compared through leave‐one‐out cross‐validation. In order to predict soil variables by different models, the predictors of OLS and REML regressions were obtained from principal component analysis (PCA), PLSR and the minimum noise fraction (MNF) transformations of spectral data on bare soil or vegetation images. The PLSR did not provide satisfactory results in terms of root mean square error (RMSE) and ratio of performance to interquartile range (RPIQ) statistics, even with hyperspectral data, mainly because of the poor signal to noise ratio (SNR) of the Hyperion sensor. The estimation accuracy increased by using the MNF method in combination with a linear mixed effect model. A multivariate approach was sometimes better than univariate ordinary kriging (OK), demonstrating the value of including Hyperion bare soil or vegetation data in the estimation procedure. Hyperspectral data provided better results than multispectral data for clay, sand and especially for SOM estimation, highlighting the value of high‐resolution spectral data for soil‐related applications.

Reducing the Influence of Soil Moisture on the Estimation of Clay from Hyperspectral Data: A Case Study Using Simulated PRISMA Data

“…Only Hyperion (on board of the EO-1 platform of NASA, USA) has a comparable spectral and spatial resolution, however, owing to the low SNR ratio of this sensor between 1900 and 2500 nm [41], the estimation of clay content is strongly hampered. Using Hyperion data, [42] obtained similar RPIQ values to those shown in this paper (i.e., 2.40) when employing spatial techniques (i.e., linear mixed effect models), while when using PLSR, the RPIQ was drastically reduced to 1.6. Accordingly, the soil clay content estimation from simulated PRISMA data could be improved by using predictive models that exploit the spatial correlation between ground samples Table 4; (b) the spectral clay index used the one is reported in Table 4 for class D.…”

Prediction of Topsoil Organic Carbon Using Airborne and Satellite Hyperspectral Imagery

“…Their results using both SMLR and artificial neural networks were significantly better than those using only SMLR or PLSR, but their models showed a low prediction ability, with RPD = 0.68. Castaldi et al [42] used Hyperion data to estimate organic matter in maize crops and did not find significant differences between the validation results obtained under bare soil and vegetation cover conditions. They obtained PLSR models with low prediction ability (RPD = 1.02) under vegetation conditions, using a calibration dataset with organic matter in the range of 1.1-2.7%.…”

Soil Organic Carbon Estimation in Croplands by Hyperspectral Remote APEX Data Using the LUCAS Topsoil Database

“…Unlike airborne data, satellite acquisition ensures a cheap way to obtain spectral data over very large areas. However, the quantitative prediction of soil properties using the first generation of hyperspectral satellite sensors is hampered by the very low signal-to-noise ratio (SNR) in the SWIR region for Hyperion imagers on board of the NASA EO-1 platform [9,18,19], or by the restricted spectral range (415-1050 nm) for the Compact High Resolution Imaging Spectrometer (CHRIS) on the European Space Agency's PROBA platform [9,[20][21][22]. Because of the low SNR, Zhang et al [23] obtained unreliable PLSR predictions for soil moisture and clay content using Hyperion image reflectance spectra.…”

“…Because of the low SNR, Zhang et al [23] obtained unreliable PLSR predictions for soil moisture and clay content using Hyperion image reflectance spectra. In order to decrease the SNR of the Hyperion data, Castaldi et al [19] used the bands retrieved from the minimum-noise fraction transformation (MNF) as regressors in the multivariate models for the clay, sand and organic matter estimation; in this case, the authors managed to improve the estimation accuracy as compared to multispectral data obtained by the Advance Land Imager (ALI). In the near future, at least five satellites equipped with hyperspectral imagers are due to be launched: the German Environmental Mapping and Analysis Program (EnMap) [24], the Italian PRecursore IperSpettrale della Missione Applicativa (PRISMA) [25], the U.S. NASA Hyperspectral Infrared Imager (HyspIRI) [26], the Japanese Hyperspectral Imager Suite (HISUI) [27], the Israeli Hyperspectral imager (SHALOM) [28], and the China Commercial Remote-sensing Satellite System (CCRSS).…”

“…Most authors directly calibrate the airborne or satellite spectra on soil samples collected from the corresponding pixels [16,19,[21][22][23]29]. This involves collection of soil samples within bare fields, laboratory measurement of the target variable, and calibration of a multivariate model linking the quantity of the target variable with the spectra extracted from the remote-sensing instrument at the sampling points.…”