Scholarlink

<b>Abstract</b>— <p>Based on an analysis of literature data and our own field research, we examined the feasibility of detecting eroded soils rather than erosion risk or loss intensity. Almost all classifications classify soils by degree of erosion based on fixed percentages of humus horizon erosion and/or humus content within it. Therefore, mapping of eroded soils using these classifications, as well as the <i>Soil Taxonomy</i> and <i>WRB</i> classifications, is not feasible. One method for detecting eroded soils is to analyze images of the exposed surface of arable soils using remote sensing data. Factors such as the properties of the plow horizon, weather conditions, and the presence of weeds and plant debris on the surface collectively shape the spectral reflectance of the arable soil surface and, consequently, the characteristics of its representation on remote sensing data. This diversity of factors, given their dynamic nature, makes the task of accurately identifying classification units of eroded soils difficult to achieve. This is only possible with a large degree of error in establishing units’ boundaries, and not for all soils and not for all degrees of erosion. The feasibility of approximately identifying areas depends on the specific soil patterns of the study area, the characteristics of the remote sensing data used, the time of their acquisition, spatial resolution, and the weather conditions prior to acquisition. Areas of strongly eroded soils can be more reliably (albeit approximately) identified using remote sensing data; units of moderately eroded soils are less reliably identified; and only in rare cases it is possible to identify slightly eroded soils. In the future, accurate identification of eroded soils can be achieved via combining automated remote sensing analysis and computer modeling of soil loss, but not within the erosion gradations established in soil classification. A new classification of eroded soils based on remote sensing data analysis and computer modeling methods is needed.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Detection of Eroded Arable Soils Based on Remote Sensing Data: Potential and Limitations

  • 期刊论文

  • 发表日期：

    2026年6月22日

查看全文
  • I. Yu. Savin

摘要

Abstract—

Based on an analysis of literature data and our own field research, we examined the feasibility of detecting eroded soils rather than erosion risk or loss intensity. Almost all classifications classify soils by degree of erosion based on fixed percentages of humus horizon erosion and/or humus content within it. Therefore, mapping of eroded soils using these classifications, as well as the Soil Taxonomy and WRB classifications, is not feasible. One method for detecting eroded soils is to analyze images of the exposed surface of arable soils using remote sensing data. Factors such as the properties of the plow horizon, weather conditions, and the presence of weeds and plant debris on the surface collectively shape the spectral reflectance of the arable soil surface and, consequently, the characteristics of its representation on remote sensing data. This diversity of factors, given their dynamic nature, makes the task of accurately identifying classification units of eroded soils difficult to achieve. This is only possible with a large degree of error in establishing units’ boundaries, and not for all soils and not for all degrees of erosion. The feasibility of approximately identifying areas depends on the specific soil patterns of the study area, the characteristics of the remote sensing data used, the time of their acquisition, spatial resolution, and the weather conditions prior to acquisition. Areas of strongly eroded soils can be more reliably (albeit approximately) identified using remote sensing data; units of moderately eroded soils are less reliably identified; and only in rare cases it is possible to identify slightly eroded soils. In the future, accurate identification of eroded soils can be achieved via combining automated remote sensing analysis and computer modeling of soil loss, but not within the erosion gradations established in soil classification. A new classification of eroded soils based on remote sensing data analysis and computer modeling methods is needed.