Scholarlink

<p>Soil organic matter (SOM) is a key indicator of soil health in cities. However, its monitoring in heterogeneous urban landscapes is still constrained by the cost and time requirements of conventional analyses. Visible and near-infrared (Vis-NIR) spectroscopy offers a rapid, non-destructive, and cost-effective alternative, but its use in urban soils and its integration with emerging measurement standards remain limited. In this study, we evaluated the ability of Vis–NIR spectroscopy to predict SOM content in 150 topsoil samples (0–20&#xa0;cm) collected from seven contrasting urban land-use systems (LUS) in Zagreb, Croatia. SOM content ranged from 1.67 to 12.6% (mean 4.35%), with statistically significant differences among LUS (<i>p</i> ≤ 0.05), where school playgrounds and public green areas next to roads showed up to 38% higher SOM than kindergarten playgrounds. Reflectance spectra (350–2500&#xa0;nm) were acquired following the AI4SoilHealth / IEEE P4005-aligned protocol and pre-processed using band clipping, Savitzky–Golay smoothing and standard normal variate (SNV). A partial least squares regression (PLSR) model calibrated on 80% of the samples and independently validated on the remaining 20% achieved excellent predictive performance (R² = 0.96, RMSE = 0.35%). Spectral signatures varied with SOM content and LUS, with characteristic absorption features identified in the VIS and NIR regions. Wavelengths with the highest regression coefficients were found around 455&#xa0;nm, 853&#xa0;nm and 1916–2236&#xa0;nm. Our results demonstrate that, even in a small but highly heterogeneous urban area, a city-specific spectral library combined with standardized measurement protocols can provide robust SOM estimates. This work advances the application of soil spectroscopy to urban environments and supports the integration of Vis–NIR methods into urban soil health monitoring and planning frameworks.</p> Graphical abstract <p></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

Spectroscopic assessment of soil organic matter in urban soils under diverse land use systems

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月26日

查看全文
  • Sabina Poudel,
  • Martin Kulhanek,
  • Helena Bakić Begić,
  • Jelena Horvatinec,
  • Marko Reljić,
  • Monika Zovko

摘要

Soil organic matter (SOM) is a key indicator of soil health in cities. However, its monitoring in heterogeneous urban landscapes is still constrained by the cost and time requirements of conventional analyses. Visible and near-infrared (Vis-NIR) spectroscopy offers a rapid, non-destructive, and cost-effective alternative, but its use in urban soils and its integration with emerging measurement standards remain limited. In this study, we evaluated the ability of Vis–NIR spectroscopy to predict SOM content in 150 topsoil samples (0–20 cm) collected from seven contrasting urban land-use systems (LUS) in Zagreb, Croatia. SOM content ranged from 1.67 to 12.6% (mean 4.35%), with statistically significant differences among LUS (p ≤ 0.05), where school playgrounds and public green areas next to roads showed up to 38% higher SOM than kindergarten playgrounds. Reflectance spectra (350–2500 nm) were acquired following the AI4SoilHealth / IEEE P4005-aligned protocol and pre-processed using band clipping, Savitzky–Golay smoothing and standard normal variate (SNV). A partial least squares regression (PLSR) model calibrated on 80% of the samples and independently validated on the remaining 20% achieved excellent predictive performance (R² = 0.96, RMSE = 0.35%). Spectral signatures varied with SOM content and LUS, with characteristic absorption features identified in the VIS and NIR regions. Wavelengths with the highest regression coefficients were found around 455 nm, 853 nm and 1916–2236 nm. Our results demonstrate that, even in a small but highly heterogeneous urban area, a city-specific spectral library combined with standardized measurement protocols can provide robust SOM estimates. This work advances the application of soil spectroscopy to urban environments and supports the integration of Vis–NIR methods into urban soil health monitoring and planning frameworks.

Graphical abstract