Scholarlink

Surface Soil Moisture (SSM) is a fundamental information in various research contexts. The monitoring of soil moisture variations can be performed through an accurate use of sensors calibrated and distributed over the study areas. Also, remote sensing techniques can be used to estimate SSM, analyzing wide areas automatically. Starting from previous studies about the response of capacitive soil moisture sensors (WaterScout SM100) installed in the study area of Ceriana-Mainardo (Liguria, Italy), the present work aims to investigate the correlation between ground-based soil moisture data and reflectance extracted from Sentinel-2 images in a study area characterized by sparsely vegetated surfaces. Despite S-2 limitations due to sensitivity to atmospheric obstacles, they provide very effective monitoring of vegetation conditions, thanks to the Red Edge bands. In fact, Reflectance – Soil Moisture correlation analysis highlighted a better response for Red Edge 2, Red Edge 3, Broad Near Infrared and Near Infrared bands, being the vegetation cover spectral response an indirect indicator of soil moisture in the investigated depths. The best correlation is found referring to soil moisture at 10 cm depth and considering the mean reflectance of the four mentioned bands. This kind of dependency allows to have a quite good correlation (R2 = 0.56, RES STD = 3.3% θ (m3/m3), MAE = 2.1% θ (m3/m3)), comparable with the accuracy of the ground soil moisture sensors (3% θ), hence useful to extract spatially distributed information of Volumetric Water Content (θ [m3/m3%]) from S-2 images. The results of the analysis appear encouraging. The workflow, implemented on an area mostly characterized by complex cultivations, could be replicated on other vegetated land uses in presence of field soil moisture data in the study area.

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

Surface Soil Moisture Estimate by Integration of Optical Remote Sensing and Low-Cost Field Sensor Network: The Case Study of Ceriana-Mainardo (Liguria, Italy)

  • 会议论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年

查看全文
  • Alessandro Iacopino,
  • Rossella Bovolenta,
  • Bianca Federici

摘要

Surface Soil Moisture (SSM) is a fundamental information in various research contexts. The monitoring of soil moisture variations can be performed through an accurate use of sensors calibrated and distributed over the study areas. Also, remote sensing techniques can be used to estimate SSM, analyzing wide areas automatically. Starting from previous studies about the response of capacitive soil moisture sensors (WaterScout SM100) installed in the study area of Ceriana-Mainardo (Liguria, Italy), the present work aims to investigate the correlation between ground-based soil moisture data and reflectance extracted from Sentinel-2 images in a study area characterized by sparsely vegetated surfaces. Despite S-2 limitations due to sensitivity to atmospheric obstacles, they provide very effective monitoring of vegetation conditions, thanks to the Red Edge bands. In fact, Reflectance – Soil Moisture correlation analysis highlighted a better response for Red Edge 2, Red Edge 3, Broad Near Infrared and Near Infrared bands, being the vegetation cover spectral response an indirect indicator of soil moisture in the investigated depths. The best correlation is found referring to soil moisture at 10 cm depth and considering the mean reflectance of the four mentioned bands. This kind of dependency allows to have a quite good correlation (R2 = 0.56, RES STD = 3.3% θ (m3/m3), MAE = 2.1% θ (m3/m3)), comparable with the accuracy of the ground soil moisture sensors (3% θ), hence useful to extract spatially distributed information of Volumetric Water Content (θ [m3/m3%]) from S-2 images. The results of the analysis appear encouraging. The workflow, implemented on an area mostly characterized by complex cultivations, could be replicated on other vegetated land uses in presence of field soil moisture data in the study area.