Mapping Soil Salinity/Sodicity by using Landsat OLI Imagery and PLSR Algorithm over Semiarid West Jilin Province, China

Authors: Mingyue Liu, Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Hao YU*, , Baojia Du, Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Zongming Wang, Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Liangjun Hu, Northeast Normal University, Key Laboratory for Vegetation Ecology Science of Ministry of Education, , Bai Zhang, Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences
Topics: Applied Geography, Arid Regions, China
Keywords: soil salinity, soil sodicity, Partial Least Square Regression (PLSR), hybridized salinity and sodicity (HSS), Landsat 8 OLI
Session Type: Paper
Day: 4/5/2019
Start / End Time: 3:05 PM / 4:45 PM
Room: Jackson, Marriott, Mezzanine Level
Presentation File: No File Uploaded


Soil salinity and sodicity can significantly reduce the value and the productivity of affected lands, posing degradation, and threats to sustainable development of natural resources on earth. This research attempted to map soil salinity/sodicity via disentangling the relationships between Landsat 8 Operational Land Imager (OLI) imagery and in-situ measurements (EC, pH) over the west Jilin of China. We established the retrieval models for soil salinity and sodicity using Partial Least Square Regression (PLSR). Spatial distribution of the soils that were subjected to hybridized salinity and sodicity (HSS) was obtained by overlay analysis using maps of soil salinity and sodicity in geographical information system (GIS) environment. We analyzed the severity and occurring sizes of soil salinity, sodicity, and HSS with regard to specified soil types and land cover. Results indicated that the models’ accuracy was improved by combining the reflectance bands and spectral indices that were mathematically transformed. Therefore, our results stipulated that the OLI imagery and PLSR method applied to mapping soil salinity and sodicity in the region. The mapping results revealed that the areas of soil salinity, sodicity, and HSS were 1.61 × 106 hm2, 1.46 × 106 hm2, and 1.36 × 106 hm2, respectively. Also, the occurring area of moderate and intensive sodicity was larger than that of salinity. This research may underpin efficiently mapping regional salinity/sodicity occurrences, understanding the linkages between spectral reflectance and ground measurements of soil salinity and sodicity, and provide tools for soil salinity monitoring and the sustainable utilization of land resources.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login