No abstract

Missing data is a common issue in remote sensing. Data reconstruction through multiple satellite data sources has become one of the most powerful ways to solve this issue. Continuous monitoring of suspended particulate matter (SPM) in arid lakes is vital for water quality solutions. Therefore, this research aimed to develop and evaluate the performance of two image reconstruction strategies, spatio-temporal fusion reflectance image inversion SPM and SPM spatio-temporal fusion, based on the measured SPM concentration data with Sentinel-2 and Sentinel-3. The results show that (1) ESTARFM (Enhanced Spatio-temporal Adaptive Reflection Fusion Model) performed better than FSDAF (Flexible Spatio-temporal Data Fusion) in the fusion image generation, particularly the red band, followed by the blue, green, and NIR (near-infrared) bands. (2) A single-band linear and non-linear regression model was constructed based on Sentinel-2 and Sentinel-3. Analysis of the accuracy and stability of the model led us to the conclusion that the red band model performs well, is fast to model, and has a wide range of applications (Sentinel-2, Sentinel-3, and fused high-accuracy images). (3) By comparing the two data reconstruction strategies of spatio-temporal fused image inversion SPM and spatio-temporal fused SPM concentration map, we found that the fused SPM concentration map is more effective and more stable when applied to multiple fused images. The findings can provide an important scientific reference value for further expanding the inversion research of other water quality parameters in the future and provide a theoretical basis as well as technical support for the scientific management of Ebinur Lake’s ecology and environment.

Ebinur Lake is a shallow lake and vulnerable to strong winds, which can lead to drastic changes in suspended particulate matter (SPM). High spatial and temporal resolution images are therefore urgently needed for SPM monitoring over the Ebinur Lake. Hence, a high-efficiency inversion model of estimating SPM from high-resolution images using machine learning is essential to increase the amount of extracted information through band combinations quadratic optimization. This article aims to evaluate the capability of the PlanetScope images and four machine learning approaches for estimating SPM of the Ebinur Lake. The specific objectives include: to obtain the sensitive bands and band combinations for SPM using correlation analysis; to quadratically optimize the combination pattern of sensitive bands using a linear model; and to compare the accuracy of traditional linear model and machine learning models in estimating SPM. The results of the study confirm that after linear model quadratic optimization, the band combinations of B3 * B4, (B2+B3)/ (B2-B3), (B3+B4) * (B3+B4), and (B3-B2)/(B2/B3) have higher accuracy than that of the single band model. By inputting the preferred four-band combinations into the partial least squares, random forest, extreme gradient boosting, gradient boosting decision tree, and categorical boosting (CatBoost) models, the performance of the SPM inversion based on PlanetScope images is better than the traditional linear model. Validation of the inversion maps with observations further indicates that the CatBoost model performed the best.