Süzer et al. (2025) Remote Screening of Nitrogen Uptake and Biomass Formation in Irrigated and Rainfed Wheat

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Nitrogen
• Year: 2025
• Date: 2025-09-09
• Authors: Mehmet Hadi Süzer, Ferit Kiray, Emrah Ramazanoğlu, Mehmet Alı Çullu, Nusret Mutlu, Ahmet Yılmaz, Roland Bol, Mehmet Şenbayram
• DOI: 10.3390/nitrogen6030082

Research Groups

Not specified in the provided text.

Short Summary

This study evaluated the effectiveness of drone- and satellite-based spectral indices combined with neural network models for estimating wheat biomass and nitrogen uptake in water-limited environments, finding that while indices predict biomass well, accurate nitrogen uptake estimation requires integrating them with complementary crop traits in nonlinear models.

Objective

• To evaluate the effectiveness of drone- and satellite-based spectral indices, in combination with neural network models, for estimating biomass and nitrogen uptake in rainfed and irrigated wheat under water-limited conditions.

Study Configuration

• Spatial Scale: Large-scale field experiments in South-East Turkey.
• Temporal Scale: Throughout the vegetation period of wheat (one growing season).

Methodology and Data

• Models used: Two-layer feed-forward neural network model.
• Data sources:
  • Drone-based spectral indices (e.g., CLRed_edge).
  • Satellite-based spectral indices.
  • SPAD values (chlorophyll content).
  • Plant height values.
  • Field experiments with varying nitrogen fertilizer rates:
    • Irrigated fields: 0, 60, 120, and 160 kg N ha⁻¹.
    • Rainfed fields: 0, 20, 40, 50, and 60 kg N ha⁻¹.
  • Fresh biomass and grain yield measurements.

Main Results

• Highest fresh biomass was 57.7 ± 1.1 tonnes per hectare for irrigated treatments and 15.9 ± 1.0 tonnes per hectare for rainfed treatments.
• Grain yield was 2.5-fold higher in irrigated wheat (8.2 ± 1.2 tonnes per hectare) compared to rainfed wheat (2.9 ± 0.9 tonnes per hectare).
• Drone-based spectral indices, particularly those based on the red-edge region (CLRed_edge), correlated strongly with biomass (R² > 0.9 in irrigated wheat).
• Spectral indices alone were insufficient to explain crop nitrogen concentration throughout the vegetation period, attributed to the nitrogen dilution effect.
• Integrating spectral indices with supplementary input parameters (SPAD and plant height) using a two-layer feed-forward neural network model substantially enhanced predictions of nitrogen uptake (R² up to 0.95).
• A simplified neural network model using only NDVI and plant height parameters still achieved significant performance for nitrogen uptake estimation (R² = 0.84).
• Overall, spectral indices are reliable predictors of biomass but require integration with complementary crop traits in nonlinear models for acceptable estimates of nitrogen uptake.

Contributions

• Demonstrated that while remote sensing spectral indices are effective for estimating crop biomass, they are insufficient for accurate nitrogen concentration or uptake estimation in water-limited environments.
• Showed that integrating spectral indices with complementary crop traits (e.g., SPAD, plant height) using nonlinear models (e.g., neural networks) significantly improves the accuracy of nitrogen uptake predictions.
• Provided a practical approach for more precise fertilizer management and sustainable wheat production under water-limited conditions by leveraging combined remote sensing and in-situ data with advanced modeling.

Funding

Not specified in the provided text.

Citation

@article{Süzer2025Remote,
  author = {Süzer, Mehmet Hadi and Kiray, Ferit and Ramazanoğlu, Emrah and Çullu, Mehmet Alı and Mutlu, Nusret and Yılmaz, Ahmet and Bol, Roland and Şenbayram, Mehmet},
  title = {Remote Screening of Nitrogen Uptake and Biomass Formation in Irrigated and Rainfed Wheat},
  journal = {Nitrogen},
  year = {2025},
  doi = {10.3390/nitrogen6030082},
  url = {https://doi.org/10.3390/nitrogen6030082}
}