Scholarlink

<p>Nitrate continues to be a major threat to drinking water resources, but rapid changes in concentrations cannot be addressed by standard laboratory approaches. This study introduces a low-cost optical sensor for real-time, in-situ monitoring of nitrate (NO<sub>3</sub><sup>−</sup>) and dissolved organic carbon (DOC) concentrations in natural water samples (soil water, groundwater and river water). Utilizing absorbance and fluorescence at specific wavelengths with LEDs and photodiodes, this sensor system offers an alternative to expensive and complex laboratory or in-situ spectrometer methods and is suited to be paired with flux measurements (e.g., lysimeters) to assess trends and dynamics. Rather than relying on costly xenon lamps and spectrometers, and therefore external power supply, our system consists of three modules that use only LEDs and photodiodes and are optimized for detection at the specific UVA, UVC and red wavelengths. This configuration enables measurements on a broad variety of samples including laboratory standards, groundwater, stream water, and soil water extracts. Initial tests with laboratory nitrate standard solutions up to 100 mg/l achieved high accuracy, with a linear model exhibiting an R<sup>2</sup> of 0.99 and mean absolute error (MAE—average magnitude of errors in a set of predictions) of 2.63 mg/l. Although the sensor's accuracy does not fully match that of traditional laboratory analyses like ion chromatography or photogrammetric approaches, it maintains good predictive capabilities with R<sup>2</sup> values exceeding 0.9 and MAE of 4.2 mg/l NO<sub>3</sub><sup>−</sup> for a sample mixture of groundwater, stream water and soil water with concentrations up to 66 mg/l NO<sub>3</sub><sup>−</sup>. DOC can be predicted with a MAE of 2.2 mg/l. Challenges such as the interference of DOC and turbidity with the nitrate absorbance signal, intense calibration procedures and site-specific variability remain, necessitating further refinement. Nevertheless, this sensor system provides a significant step toward accessible, continuous water quality monitoring and lays the foundation for linking nitrate concentrations to in-situ fluxes. These advancements are crucial for enhancing nutrient management and environmental protection practices.</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

Developing a low-cost nitrate and DOC sensor for natural water samples

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月7日

查看全文
  • Heinke Paulsen,
  • Christof Hübner,
  • Markus Weiler

摘要

Nitrate continues to be a major threat to drinking water resources, but rapid changes in concentrations cannot be addressed by standard laboratory approaches. This study introduces a low-cost optical sensor for real-time, in-situ monitoring of nitrate (NO3−) and dissolved organic carbon (DOC) concentrations in natural water samples (soil water, groundwater and river water). Utilizing absorbance and fluorescence at specific wavelengths with LEDs and photodiodes, this sensor system offers an alternative to expensive and complex laboratory or in-situ spectrometer methods and is suited to be paired with flux measurements (e.g., lysimeters) to assess trends and dynamics. Rather than relying on costly xenon lamps and spectrometers, and therefore external power supply, our system consists of three modules that use only LEDs and photodiodes and are optimized for detection at the specific UVA, UVC and red wavelengths. This configuration enables measurements on a broad variety of samples including laboratory standards, groundwater, stream water, and soil water extracts. Initial tests with laboratory nitrate standard solutions up to 100 mg/l achieved high accuracy, with a linear model exhibiting an R2 of 0.99 and mean absolute error (MAE—average magnitude of errors in a set of predictions) of 2.63 mg/l. Although the sensor's accuracy does not fully match that of traditional laboratory analyses like ion chromatography or photogrammetric approaches, it maintains good predictive capabilities with R2 values exceeding 0.9 and MAE of 4.2 mg/l NO3− for a sample mixture of groundwater, stream water and soil water with concentrations up to 66 mg/l NO3−. DOC can be predicted with a MAE of 2.2 mg/l. Challenges such as the interference of DOC and turbidity with the nitrate absorbance signal, intense calibration procedures and site-specific variability remain, necessitating further refinement. Nevertheless, this sensor system provides a significant step toward accessible, continuous water quality monitoring and lays the foundation for linking nitrate concentrations to in-situ fluxes. These advancements are crucial for enhancing nutrient management and environmental protection practices.