Abstract
Accurate digital mapping of soil organic carbon (SOC) is important in understanding the global carbon cycle and its implications in mitigating climate change. Visible and near-infrared hyperspectral imaging technology provides an alternative for mapping SOC efficiently and accurately, especially at regional and global scales. However, there is a lack of understanding of the impacts of spatial resolution of hyperspectral images and spatial autocorrelation of spectral information on the accuracy of SOC retrievals. In this study, the hyperspectral images (380–1700 nm) with a spatial resolution of 1 m were acquired by Headwall Micro-Hyperspec airborne sensors. Then, hyperspectral images were resampled into three different spatial resolutions of 10 m, 30 m, and 60 m by near neighbor (NN), bilinear interpolation (BI), and cubic convolution (CC) resampling methods. The geographically weighted regression (GWR) model was used to explore the role of spatial autocorrelation in predicting SOC contrast with the partial least squares regression (PLSR) model. Results showed that (1) the hyperspectral images can be used to predict SOC and the spatial autocorrelation can improve the prediction accuracy, as the ratio of performance to interquartile range (RPIQ) values of PLSR and GWR were 1.957 and 2.003; (2) The SOC prediction accuracy decreased with the degradation of spatial resolution, and the RPIQ values of PLSR were from 1.957 to 1.134, and of GWR were from 2.003 to 1.136; (3) Three resampling methods had a much weaker influence than spatial resolution on SOC predictions because the differences of RPIQ values of NN, BI, and CC resampling methods were 0.146, 0.175, and 0.025 in the spatial resolutions of 10 m, 30 m, and 60 m, respectively; (4) Finally, the Global Moran’s I and the Anselin Local Moran’s I proved the existence of the spatial autocorrelation in SOC maps. We hope that this study can offer valuable information for digital soil mapping by satellite hyperspectral images in the near future.
Highlights
Soil holds the largest terrestrial organic carbon stock, which is approximately three times more than the vegetation C-pool and approximately twice that of the atmospheric C-pool [1]
The objectives of this study were to (1) evaluate the prediction accuracy of hyperspectral remote sensing (HRS) imaging in mapping soil organic carbon (SOC) by partial least squares regression (PLSR) and geographically weighted regression (GWR), (2) explore the impact of spatial resolution on SOC prediction accuracy, and (3) determine the role of spatial autocorrelation in predicting SOC
The dispersion degree of the SOC was small in this study region, and this can further help evaluate the predictive ability of the models
Summary
Soil holds the largest terrestrial organic carbon stock, which is approximately three times more than the vegetation C-pool and approximately twice that of the atmospheric C-pool [1]. The conventional approach for mapping SOC relies on field samplings and wet chemical analyses in the laboratory (e.g., potassium dichromate), which are time-consuming, labor-intensive, and prohibitively expensive [3,4,5]. Laboratory-visible and near-infrared (VNIR) spectra have been successfully used to estimate SOC in many studies [12,13,14,15]. There is a tendency to avoid these labor-extensive steps by collecting in situ VNIR spectra to achieve more rapid SOC estimations [16,17]. The efficiency of SOC estimations by laboratory or in situ hyperspectral imaging is higher than that of traditional methods, sampling points are still sparsely distributed in study regions. Spatially continuous SOC maps cannot be obtained by these discrete soil samples
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
