Article
Agriculture, Multidisciplinary
Congcong Lao, Junying Chen, Zhitao Zhang, Yinwen Chen, Yu Ma, Haorui Chen, Xiaobo Gu, Jifeng Ning, Jiming Jin, Xianwen Li
Summary: This study aimed to enhance the potential of visible-near infrared (VIS-NIR) spectroscopy in predicting salt and major soluble ions in topsoil in Inner Mongolia. By using fractional-order derivatives and spectral parameters for model calibration, the optimal prediction accuracy varied across different ions, with models exceeding quantitative prediction levels.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2021)
Article
Environmental Sciences
Kangying Zhu, Zhigang Sun, Fenghua Zhao, Ting Yang, Zhenrong Tian, Jianbin Lai, Wanxue Zhu, Buju Long
Summary: Abundant shallow underground brackish water resources in the North China Plain could help alleviate the shortage of fresh water resources and crisis concerning agricultural water resources. However, improper brackish water irrigation may increase soil salinity and decrease crop yield. This study found that optimized SVIs, including soil salt-sensitive wavebands, performed better in retrieving soil salinity, especially at the 30-cm depth, providing a practical technique for evaluating regional brackish water irrigation systems.
Article
Geochemistry & Geophysics
Munmun Baisantry, Anil Kumar Sao, Dericks Praise Shukla
Summary: In this paper, a supervised spectral-spatial band selection method is proposed, which combines the advantages of dimensionality reduction and feature extraction to improve classification accuracy.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Multidisciplinary Sciences
Jiawen Hou, Yusufujiang Rusuli
Summary: Low-cost and efficient dynamic monitoring of surface salinization information is crucial in arid and semi-arid regions. This study conducted a remote sensing inversion exercise in the Bosten Lake watershed, Xinjiang, China, using Sentinel MSI and Landsat OLI data combined with measured soil salinity data. A grid-search support vector machine (GS-SVM) model was created to estimate soil salt content, and the optimal model was found to have a prediction ability of R-2=0.64 and RMSE=3.12. The study revealed the severe salinization and distribution characteristics of soil salinity in the Bosten Lake watershed, providing a scientific basis for further soil salinization monitoring in arid areas.
Article
Environmental Sciences
Yin Wang, Chengxiao Hu, Xu Wang, Guangyu Shi, Zheng Lei, Yanni Tang, Huan Zhang, Hada Wuriyanghan, Xiaohu Zhao
Summary: Soil salinization negatively affects soybean production, but rhizosphere microorganisms can improve plant salt tolerance. Selenium is known to optimize the rhizosphere microbial community, and this study investigated whether selenium-induced rhizosphere microorganisms can enhance plant salt tolerance. Pot experiments were conducted using salt-tolerant and salt-sensitive soybean varieties, and the results showed that selenium application improved soybean salt tolerance by optimizing the structure of the rhizosphere microbial community. Furthermore, the application of four salt-tolerant bacteria isolated from selenium-fertilized soil led to significant increases in plant growth and reductions in stress-related compounds in salt-sensitive soybean.
ENVIRONMENTAL RESEARCH
(2023)
Article
Computer Science, Artificial Intelligence
Jun Wang, Chang Tang, Xiao Zheng, Xinwang Liu, Wei Zhang, En Zhu
Summary: This paper proposes a graph regularized spatial-spectral subspace clustering method (GRSC) for hyperspectral band selection. The method preserves the spatial information of hyperspectral images through superpixel segmentation and generates discriminative latent features to represent the bands. It explores spectral correlation using a self representation subspace clustering model and regularization, and learns a similarity graph between region-aware latent features to preserve the spatial structure of the images.
Article
Environmental Sciences
Libing Wang, Bo Zhang, Qian Shen, Yue Yao, Shengyin Zhang, Huaidong Wei, Rongpeng Yao, Yaowen Zhang
Summary: This study evaluated the performance of hyperspectral data in estimating soil salt content, showing that satellite hyperspectral data was superior to laboratory spectral data. In the modeling process, raw spectra performed the best, with high R-2 values and low errors in estimating sodium ion and salt content, while first derivative analysis and principal component analysis performed better in estimating chloride and sulfate content.
Article
Environmental Sciences
Elizabeth Baby George, Cecile Gomez, D. Nagesh Kumar, Subramanian Dharumarajan, Manickam Lalitha
Summary: Hyperspectral imaging spectroscopy is a useful tool for mapping soil properties at large scales. This study analyzed the impact of bare soil pixel identification on clay content estimation using two methods: spectral indices and spectral unmixing. The results showed that the spectral unmixing method provided slightly better performances in estimating clay content, although it reduced the spatial coverage.
GEOCARTO INTERNATIONAL
(2022)
Article
Limnology
Christopher T. Solomon, Hilary A. Dugan, William D. Hintz, Stuart E. Jones
Summary: The widespread and increasing use of road deicing salt is causing an increase in lake chloride concentrations, which has negative impacts on aquatic organisms and ecosystems. A simple model was used to study the factors affecting road salt concentrations and predict equilibrium concentrations in lakes across the contiguous United States. The model shows that equilibrium salt concentration depends on salt application rate, road density, and runoff. By controlling or reducing salt application rates, it is possible to achieve equilibrium concentrations below recommended thresholds in many lakes. The analysis provides insights into the current trends of road salt pollution in lakes and suggests achievable goals for protecting aquatic organisms.
LIMNOLOGY AND OCEANOGRAPHY LETTERS
(2023)
Article
Environmental Studies
Eric Ariel L. Salas, Sakthi Subburayalu Kumaran
Summary: In this study, a new bare-soil index, HBSI, was developed to improve the accuracy of bare-soil remote-sensing mapping. The HBSI outperformed other existing indices and showed high classification accuracy when combined with NDVI.
Article
Agronomy
Ruiqi Du, Junying Chen, Zhitao Zhang, Yinwen Chen, Yujie He, Haoyuan Yin
Summary: The study verified the spectrum mechanism responsive to soil moisture and salinity by statistical tests, and developed the MSS model to inverse the factors affecting SR and estimate moisture and salinity accurately. The results provide dynamic information on soil moisture and salinity content and guide local irrigation management.
AGRICULTURAL WATER MANAGEMENT
(2022)
Article
Environmental Sciences
Mohamed A. E. AbdelRahman, Ahmed A. Afifi, Paola D'Antonio, Safwat S. Gabr, Antonio Scopa
Summary: This study tested previously proposed salinity indices on salt-affected soil in the northern Nile Delta region of Egypt and found that they were not suitable due to the interaction between bare soils, salts, and urbanization. To address this issue, a new index was proposed, taking into consideration plant health, salt crust, and urbanization, and it was found to be more effective in mapping salt-affected soil compared to previous indices. The accuracy of the new index was evaluated using multi-temporal satellite data and field measurements, with high accuracy results obtained.
Article
Environmental Sciences
Fatemeh Abedi, Alireza Amirian-Chakan, Mohammad Faraji, Ruhollah Taghizadeh-Mehrjardi, Ruth Kerry, Damoun Razmjoue, Thomas Scholten
Summary: The study utilized machine learning algorithms to model soil salinity, showing that remote sensing data contributed more to predicting electrical conductivity and sodium adsorption ratio, and that model averaging approach has the potential to improve prediction accuracy.
LAND DEGRADATION & DEVELOPMENT
(2021)
Article
Agronomy
Wenju Zhao, Fangfang Ma, Haiying Yu, Zhaozhao Li
Summary: This study aimed to investigate the impact of texture information and spectral index on the accuracy of soil salinity inversion models. Field data from the Bianwan Farm in Gansu Province, China were collected, and machine learning models based on different combinations of indices were constructed. The results showed that the model combining vegetation index (VI) and texture index (TI) had the best inversion effect. The Random Forest (RF) algorithm outperformed the Extreme Learning Machine (ELM) algorithm in terms of accuracy and stability. These findings provide a theoretical basis for efficient soil salinity inversion and management of saline-alkali lands.
Article
Geochemistry & Geophysics
Peng Chen, Rong Ma, Jiansheng Shi, Letian Si
Summary: Under the influences of climate change and human activities, arid inland lakes are shrinking and drying up, releasing chemical dust from the bare lake bed. This study aims to understand the impact of groundwater on vegetation distribution and development as well as soil salinization. Field investigations and analysis were conducted in Chahan Lake, a representative dry lake in northern China. The results show that groundwater depth plays a significant role in determining vegetation communities and salinization levels in the lake.